Oil Pressure and Temperature - Final

world's most dangerous man

Before we begin, my heart aches as I watch my beloved country descending into ruin. I generally try to avoid politics in this blog, but we all eventually reach a point where we can hold our silence no longer. The absence of leadership, and the propensity for the current White House occupant to throw gas on a fire are exacerbating our national catastrophe. A disbeliever attitude towards a pandemic has gotten everyone anxious as well as over 100,000 Americans killed. Add 40 million unemployed people (many of whom are not getting their promised checks) who are going hungry and are going to get upset, looking for an outlet. Add to that what appears to be police conduct inconsistencies (white folks with guns can stop ambulances from getting to the hospital, but Black and Brown folks can't just walk down the street armed with only signs). Add an elected official directing tear gas and rubber bullets at unarmed protesters so he can have a photo-shoot outside a church after threatening our citizens with military forces. The people want leadership; we want answers. After all this, we demand, and frankly deserve, a change. Okay, back to the usual chaos oasis: car stuff.

I have been slowly working through a gauge solution for the oil system in the TDI conversion (For reference, see Oil and Temperature 1, 2, 3, 4 and 5). This really has been interesting, and I think those running a stock air-cooled motor might want to do something similar. Since the air-cooled engine runs oil hotter, the broader range gauge may work better, which means less work for you. I unfortunately did not take any pictures of the final steps setting up or installing the Porsche combination gauge. My phone died, so while it sat on the charger I readied the dashpod. Then, assembly inside the bus required 3 hands so I couldn't get a picture without risking dropping something. So, please try to use your imagination as I describe the final assembly.

Final Gauge Prep - Bulbs

When I left off last, I had the new 250*F limit temperature innards swapped out, and that half of the gauge re-installed into the gauge. We were entering the home stretch. I focused on the bulbs next. There are 4: one each for illuminating the two gauges and one each for the 2 new idiot lights (Alt and Oil). Now, I don't have the light wired at the alternator, but I figured if I could get the gauge-end ready, it would be one fewer thing to fiddle with later. The illumination lights are simple push-and-twist 1.2W bulbs like the others in the bus dash. In fact, they are the exact same bulb. The idiot lights, however, are not like those, nor are they the same size. The alternator bulb is bigger, and the oil pressure light is an odd spade style. As luck would have it, back when I replaced the thermostat panel bulb in Flash, I had to buy a 2-pack of bulbs. That bulb (Sylvania 2721) was a match for the oil pressure bulb socket. Kismet!

Final Gauge Prep - Wiring
With the bulbs set, I shifted to the wiring. There are 4 signal wires (oil temp, oil pressure, oil warning and alt warning). For these, I simply stripped the wire ends. They already have disconnects at the gauge so I could disconnect them later when I pull the dashpod. The other wires (1 ground, 2 illumination and 2 switched 12V) needed a different treatment. In the interest of keeping the dashpod remove-able, I wired the 2 switched 12V source wires into a single male disconnect. I did the same for the 2 illumination wires. To the ground, I attached a female disconnect. Before I moved out to the bus, I fiddled with the gauge to make sure it was straight up/down.

In the Bus

warming up

The dashpod re-mounts similar to how it came out: slide it under the steering wheel from the center, near the stick shift, face down. Rotate the guts towards the hole. With the combo gauge in the third hole, this is a little harder, but not by much. Now, the hard part begins. I started with the illumination wire that was in the bus, cutting the end off that goes on the passenger-side end of the dashpod. I spliced in another wire with a female disconnect on the end as well as that original wire that I had just cut off. This gave an illumination signal that the combo gauge could plug into. Then, I started plugging in all the things that were there before... the grounds, fuel gauge, turn signals, and the low beam / hi beam and, of course, the illumination bulbs. Before I took any more steps, I tested things by turning the key to run, and flipping various switches to see the bulbs light up. Satisfied I had done no harm, I shifted to the new gauge wiring.

I started by inserting one of those "chair" disconnect splitters into one of the dashpod grounds and then connecting the ground wire from the combo-gauge. I then used my wire color map from my last post, heat-shrink splicing the gauge signal wires to the cable wires. Then, I plugged in the illumination male:female. Last, I sourced switched 12V with a covered female disconnect, and plugged in the male from the gauge. I thought about wiring it into the black switched circuit that powers the fuel gauge and the turn signals, but couldn't get my head around how to do it with one of those chair disconnect splitters without risking an accidental grounding. I wanted to minimize the wire cuts too. Last, I could have just put disconnects on the ends of the colored wires rather than splice into the original gauge wires; I just didn't. I still might.

thermostat just opening

The location of the new gauge interrupts the wire-flow for the brake warning light. I will need to extend those wires, or change the warning light location. Since we are still semi-restricted for CoViD-19, I won't be driving any great distance terribly soon, so I have time to resolve it. I found it interesting how much tension was on these wires, though; curious that there was no wiggle room on them. That was true when I went to remove the dashpod at the start of all this as well. Looking back, it was an issue when I went trying to figure out what was up with the fuel gauge years ago. In fact, removing the dashpod this time involved moving the wires one by one through the hole, while placing considerable tension on the wire, so I wouldn't mess up the wires during removal.

Unrelated, the couple of hours I spent in the bus getting the dashpod wired and installed I also test-ran the little heater with the new Afterburner thermostat. Once I had the temperature probe in a spot where it was not up against the outside wall, the thermostat held the interior temperature at a fairly static 70*F. Considering the wind and rain outside, it was really luxurious. Imagine how different the experience is compared to lying in the street gutter trying to get an alternator swapped out in front of that old Lake Oswego house.... in a late fall rainstorm. We've come a long way, Hapy.

Testing

after thermostat opened

With the wiring complete, I slid the dashpod into the hole. I found that the washers that are helping to hold the combo gauge to the dashpod will hang-up on the hole if you are not careful when you remove the dashpod. I will think about how to construct something to hold the gauge on better. Anyway, once roughed into place, the 4 Phillips screws will get it into the right position. Once locked down, I re-tested the system by flipping the key to run. The gauges lifted off of their bottom position just like they did on the kitchen table during the tests. This told me that the round-trip for the 2 circuits was good. Then, I turned on the lights and I could see the gauges illuminated. The oil warning light popped on, but the Alt light did not. This was all as I expected. Next, a test fire of the engine, and maybe a little cul-de-sac parade to get the gauges to move. First, of course, thread the speedo cable into the back of the speedometer.

Because of continued limp-mode issues, I ran my engine-running tests in my driveway. That made the picture-taking possible, though. The gauges respond as I would have expected. The oil pressure starts high when the engine is first started and then settles down once the engine is warm. The oil temp warms faster than the coolant. When I rev the engine up and down, the oil temp climbs and then settles. My last test was to check under the bus for fresh oil spots. There were none. You can see from the pictures what the coolant temperature is (number in lower left corner of UltraGauge).

With that, the oil system gauging is complete. I am not sure what's next. I have a cable for re-wiring the front-to-back stuff, and otherwise I'm just about running out of parts for projects. I do have some coolant hose and a stretch of aluminum pipe, though, so maybe I'll get after some part of the defroster. Of course, some day I'll have to get back to paint-prepping Zed, the 1979 280ZX. I'll figure it out when I get to the garage. Thanks, as always for following along--

Oil Pressure and Temperature (Part 5)

When I left off last time, I had a Porche combination gauge fitted to to the bus dashpod in the previously blank third-hole. I had wrangled the stock oil pressure sensor out of the oil filter housing and back into a VDO "T". Into the other side of that "T" I threaded the big dual-pole oil pressure sensor. Last, I removed a M10x1 plug near the oil pressure sensor spot and threaded in a sensor designed to capture water temperatures, topping at 150* (250*F). Today's post continues this saga with a detour into swapping out the temperature gauge innards with new guts from a new VDO gauge.

Wiring Rear

I started with the next obvious thing: wiring up the sensors in the engine bay. First was estimating where to cut the cable. I threaded it through the cable holder on the oil filter and then gave myself about a foot more. With a knife, I whittled the cable casing back to the cable holder. Then, I followed the color chart, starting with the black wire to the temperature sensor. This sensor is less than 4 inches from the cable holder, so while this was easy on the brain, it required some stretching and finger dexterity to cut, strip, attach a disconnect and heat-seal the connection. Once done, I cut the housing on the dis-connect so it could slip onto the stud on the end of the sensor. For the oil pressure related wires, I did not have ring disconnects. So, I will revisit those later, after the CoVid lockdown ends. In the meantime, I simply placed the stripped wires (red and blue for the warning light and gauge respectively) and threaded the cap down on the posts. This will work for testing purposes, but with the shaking that comes from a diesel plus the shaking of driving, one or both will fall off eventually, so I will need to apply the ring terminals.

I left the rest of the cable alone. Back when I was still suffering temperature issues, I removed the engine cover. I decided this was a great time to re-install it. So, with a couple M6 bolts and washers, the cover went back on.

More Gauge Fun

cut line on new gauge

temp gauge removed

The longer I thought about the suitability of the oil-temperature gauge in the combination cluster, the less enthusiastic I became about installing it without trying to do something about it. So, I decided I would hack-up a water temperature gauge and swap the innards with the innards of the combo gauge. My plan was for it to continue to look like the original gauge, but the top of the range would be 250*F, aligning with the sensor. This might have the opposite effect of giving me an "in the red zone" appearance when we're still down in the 230*F range. I think I can live with that, so long as the gauge gives shows a clear difference between 210*F (normal) and 235*F (my believed top end of normal).

So, how do we swap the innards? It starts with removing the temperature section from the combination gauge. It is held on with small slotted screws. Once removed, the unit pops right out. I had an extra new-old-stock (NOS), new-in-box (NIB) temperature unit I got off eBay before I got the combination gauge. I figured I would use that as my transplant target so I can revert back to the original later. I found a few differences between the original gauge innards and the new ones. I will highlight those differences as I work through this.

New VDO Gauge Destruction

old innards top
new innards bottom

At a high level, I followed the Dave Barton PDF process where you cut the housing in half to get to the inner workings of the new gauge. I found that a standard screwdriver was 1/4" thick, so I used that to mark where to cut. I threaded the collar up to that point and then traced around the gauge housing, using the collar as a guide. I taped over the gauge lens with blue painters tape and the cut the line with my hacksaw. As directed in the instructions, I cut just enough to break through and then rotate the gauge housing to extend the gap around it. The top part lifted off right where the gauge face meets the housing. Perfect instructions; nice job Mr. Barton. I also used the salad-fork method to remove the needle, using sparing force so the needle slowly lifted off the little pin. I encourage taking great care on this step.

With the needle off, I removed the small slotted screws on either side and then removed the gauge face. Then, I moved to the flip-side. With a 7mm socket, I removed the nut on the post. The tabs hold the gauge innards in by friction. So, I put the mounting collar on backwards, so I could stand the gauge on its (now missing) face, with the tabs pointing upwards. With a framing hammer, I gently encouraged the gauge innards down by tapping on the tabs and the bolt-post. If you do this take care that the pin the needle sits in does not get crushed or bent or in anyway really touched. It did not take much force to get the innards to move. I held the hammer about 4 inches from the head and tapped.

Old VDO Gauge Mount Destruction

gauge mounting plate

I started with the same process as I had done with the new VDO: salad fork on the needle. This was not as clean a process as it was with the new one. In fact, I would discourage anyone from trying to remove the needle this way if you intend to reuse the needle or the innards. In my case, the needle was damaged, and I was unable to tell how the original needle was connected to the pin. I resolved to using the new VDO needle and kept going. I removed the 2 slotted screws holding the gauge meter / face on. Underneath was a thin piece of dried up wax paper. I carefully separated that from the gauge innards.

With the gauge face torn down, I switched to the mounting side. The old gauge separates from the metal plate fairly easily. With a 7mm socket, remove the nut from the central post/bolt. With a slotted screwdriver, encourage the innards to pull away from the metal plate. It is held purely by friction.

Re-Assembly Preparation

testing old and new together

Now, we have a small pile of parts. From the new VDO innards, the tabs need to be removed from the rest of it. These simply slide out. Now, look at the 2 innards. They look very similar in some ways, but have one major difference on the not-gauge-face side. The inputs are 90* offset. If you look at the old metal mounting plate, you will see that the 4 holes around the central-post/bolt hole are 2 different sizes. The old gauge innards' holes line up with the smaller set and the new gauge innards line up with the larger set. The plastic thing that holds the tabs for the older model has 2 plastic nubs that line up with the larger holes. Those nubs hold the plastic holder and the innards stationary. If you were to cut them off, the gauge would not sit still, and could ground against the metal plate. I hadn't pieced this together when I cut them off in an attempt to get the new innards to mount. I would recommend instead that you expand the smaller set of holes so they are the same diameter as the larger set: so all 4 are the same size. Then, the plastic holder can go in in either orientation. If you hastily cut off the plastic nubs, like I did, you can solve by locating a pair on small plastic finishing tabs from one of those assemble-at-home pieces of furniture. With a little whittling, it can be formed to hold the innards in place. I suggest altering the mounting plate instead.

Re-Assembly

approximately 175*F

comparing top temp
modified on the right

However you solved for the attachment of the plastic bit to the metal plate, slide the new VDO innards onto the 2 posts. There are 2 ways the gauge innards can fit, but only one that actually works. So, when you do this, consider the orientation of the gauge so that the gauge sweep will be on the correct side. Press together firmly. Making sure the ground tab is on the bolt, thread on the 7mm nut and tighten with the socket. I considered the orientation of the new innards on the old plate and marked which side was for the signal (S) and with was for power (+) before I moved on.

If you intended to create your own gauge face, or paste on an aftermarket one, now is the time to do that. I would really like to have markings to show where important temperatures are, but I also like the simplicity of the stock gauge, so I left it alone.

To the gauge face side, I set the wax paper and then the old gauge face. I used the 2 slotted bolts from the new gauge to attach. Last, carefully press on the needle. Once it is all together, it fits back into the combination gauge housing just as easily as it was removed. If you are using the needle from the new gauge, you will need to insert a small washer between the mounting plate and the combo-gauge for each mounting bolt or the needle will hang-up on the contoured plastic near the rear (front is front) of the gauge housing. Using the 4 slotted bolts it was held on with before, mount the plate to the cluster threading each bolt through a small washer. Re-attach the wires.

Setting the Gauge Range

gauge re-installed into dashpod

As you can see from the pictures, I took the opportunity to test the new and old innards with various temperatures. First, I applied an open flame to the sensor to determine the top temperature. I then checked to see where the needles of the 2 different gauges sit when I put the sensor in boiling water. This took some doing since the "kitchen" table is 10 steps or so from kitchen-proper. I would boil some water in a small cup in the microwave and then dash over to test it. As a result, I think the picture on the above right is for around 175*F. If you consider how much gauge is left above the needle for the new innards, and that the top temp is 250*F, I will have a wide band for the temperatures we care about: 190*F to 230*F.

Well, that's an awful lot for this week. All that's left is installing it into the bus. I have to do some wire disconnects, but I don't expect it will be terribly dramatic. Now that I see the gauge in it's final form, I would absolutely recommend bus-pilots who still have their original air-cooled engine to think about installing one of these combo gauges. The original temp range is designed for the air-cooled hot oil, and it looks really good, in my humble opinion. Knowing that the new gauge innards can be swapped in eases my old concerns about long-term support as well. Just get a new VDO gauge (less than $40US today) when one of these fails, and do what I just did.

Thanks, as always for following along. Not sure what I'll work on, much less post on next. I guess I'll figure it out when I paw through my parts and incomplete projects. Stay safe and be well-

Oil Pressure and Temperature (Part 4)

This really has been a ton of fun. We have a Porsche combination gauge mounted to the rear of my removed-from-bus dashboard. We have tested the VDO sensors against it and everything seems to line up. So, today we shift to installation.

Cable, Not Wires

inside belly pan

Rather than run another whole set of individual wires, increasing the spaghetti under the bus, I went a different route. In the TDI retrospective (See TDI install retrospective: Secondary Electrical), I mentioned a cable company. Shortly after that, I decided that I would do this gauge thing and use a cable to send the signals from the engine bay to the dashboard. I figured that I had the 3 for the oil system (temp, oil pressure gauge and oil pressure idiot light). I also want to know when the coolant is low and when the electrical system is tanking, like a normal car. So, I got a 6-wire cable, leaving one open for future expansion. Similar to most of the other stuff on this project, I ordered the cable in January before things started getting CoVid-interesting.

Snake-y

harness pops out here 

I want this install to be as clean as I can make it. So, I figured I would route the cable along the same route as the original main harness. Finding where it pops out inside the front of the cab of the bus isn't very easy: it is under the windshield-washer tank. This is much more obvious from underneath, hidden inside the center belly-pan between the main rails. I drilled a matching 11/16" hole an inch or so to the driver-side of the main harness pass-through. From inside the center belly-pan section, I pushed it up through that hole, inside the bus. I then took it from the inside, along the same path as the main harness over towards the fuse box, over the ventilation tubes and up behind where the dashpod sits. I gave myself an extra foot of cable so there wouldn't be stress on it later.

Sending the cable rear-ward was the easy part. From the center belly-pan, I threaded the cable from behind the front axle along the same route as the main harness, meeting it where it emerges from the steel pipe it travels to the rear of the bus. It ran along the main harness pipe, through a small pass-through in the rear cross-beam and back towards the starter. It passes right into the engine bay where the fuel does. I coiled the extra cable there. I had ordered 30' so I had more than enough to run it exactly where I wanted it. I cut off the excess, leaving almost 10 feet of cable, so the total need was just over 20 feet. For my future self, here is the wire color -to- purpose mapping:

Circuit	Color	Purpose
1	Black	Oil Temperature Gauge
2	Red	Oil Warning Light
3	Blue	Oil Pressure Gauge
4	Orange	Coolant Warning Light
5	Yellow	Alternator Warning Light
6	Brown	open

Pressure Sensors

splitter and stock sensor in

The TDI computer already has an oil pressure signal. It comes out of the oil filter housing. I decided that if that location was good enough for the computer, it was more than good enough for my gauge. So with a 1" spanner (that was the closest I had to the required), I knocked the stock oil pressure sensor free and applied the splitter (VDO 240 850). In order to get the splitter to point the right direction (away from the oil level tube), I added a washer between the splitter and the oil filter housing. If you do this, you may need to do the same. I put the original pressure sensor into the end and the new dual sensor into the split off the side. The dual sensor has a 17mm nut close to the threads, and tightening it can only be done from below, about 30* at a time. Fun.

Temperature Sensor Stuck?
Between the insertion of the original pressure sensor and the new dual-sensor, I went after the temperature sensor. Why? I realized that once I had the dual pressure sensor in-place, I would have my access and vision blocked by it. The plug in the oil filter housing was on fairly snug, but a pop with a ratchet holding a Torx socket did the trick. I don't remember which size; my apologies for not writing it down. Once free, oil seeped out of the hole, even though I had drained the oil before I started with the pressure sensor. No matter. It just made a mess. I first tried with the longer temp sensor (VDO 323 423), but found that it hung up on something inside the housing. I was barely able to get it back out again, and after looking at the sensor after I removed it I concluded that it was not a good idea to use that sensor here. I put a picture of it down below on the right.

Temperature Sensor, Take 2

temp sensor in

I cleared the hole, and checked for depth with a small screwdriver. It seemed like the sensor should have fit, but there is no doubting the compressed look of that sensor. So, grateful I had the shorter, but lower max temp sensor, I considered the impact. Consider that this is not an air-cooled engine, whereas the Porsche combo-gauge came from one. Also, the oil and coolant exchange temperatures, cooling the hotter oil as they do and that cooler oil is what is passing through this part of the housing on it's way back to the engine. I did a quick survey of posts about oil temperatures of TDI's and even the guys who run 'em hard say their temps don't get above 230*F. In fact, most of the folks out there pose that temps above 250*F would be cooking their engines.... and these folks have a sensor in this location, so the plan is holding, so far.

them bumps aint right

So.... I decided that this sensor is probably fine, especially for the short run, while I figure out if this is the long-term solution. Key to this thinking, is to remember that the gauge is to help identify patterns and trends. Also, to remember that this is the temperature as it is re-entering the engine. So, this will be a delayed signal of a climbing temperature event. In fact, it could be masked if the coolant is effectively pulling heat out of the oil. Concluding that I'd come this far and that there was not a better port to test temperature anyway, I continued. The shorter (250*F / 120*C max) sensor threaded right in with no trouble. I lightly torqued the sensors, filled the engine with oil and called it a day.

Wait a Minute...
One nagging thought I have about this is the numbering, or lack thereof, and the red zone on the Porsche combination gauge are completely out of alignment with what I would be concerned about as I drive a TDI. These numbers ("normal" between 194*F and 248*F) are way to broad. AND, regardless of which sender I use, the red zone starts outside the top end of it (at 302*F?). So, I may need to change that gauge at some point if it is really going to be useful long term. I mean, honestly, if normal is around 210 and getting too hot is 230, I will not easily see the difference on that gauge.

Well, that's it for today. All that remains is everything else. I will work on getting the engine-compartment end of the wiring done next. Then, I'll see about getting the dashpod wired and re-installed. Depending on how exciting and time-consuming those are, it could be one or two more posts before I have oil instrumentation.

Thanks, as always, for following along. Be safe, be patient, and if you can, patronize your favorite restaurants with a go order. Now more than ever, you can vote with your wallet for the businesses you want to survive. Personally, Amazon doesn't need my money; Discount Import Parts (now only on the east side) and the local mom-and-pop grocery do.

Oil Pressure and Temperature (Part 3)

Hapy cinco-de-Mayo. My last post ended with some testing of the Porsche combination gauge with the modern VDO sensors. I'll pick up where I left off.

Where Were We?

Porsche combo gauge

After I wrote all of those posts about the TDI install, I realized that the little UltraGauge does not show my oil temperature or pressure. I can see the coolant temperature, though. When we drive up a long grade, we can see the coolant temp climb, and I wondered if the oil temperature was leading that charge upward, dumping heat into the oil:coolant temperature exchanger. So, I started collecting sensors to compliment the gauges I had lying around from back when I had the original air-cooled engine and thought that having an effectively oil-cooled engine without a gauge about oil was just silly. That's especially true when that engine was 5 meters away. In my usual fashion, I can't just run wires from sensors up to the dash and stick a couple gauges in back of my windscreen and call it good. Oh no. That would just be too easy. Instead, I got this notion that a combination gauge from a 70's-era Porsche could fit into the 3rd hole in a VW bus dashpod. I was right. But, does it work with the new VDO sensors? Why, yes it does.

Simple Initial Testing
I did some very simple tests of the sensor-to-gauge circuit in my last post. All I did was demonstrate that the gauge would respond to 12V, and if I had a sensor hooked up, the gauge would lift slightly off the bottom post. This is no casual "who cares" moment, in my opinion. Most important, this indicated to me that the big circular thing that I had shipped halfway around the world had operational gauges. But, would they respond correctly to real-world input from modern VDO sensors?

Temperature

water still steaming, needle at 9:00

Testing the viability of the oil temperature circuit was fairly easy, once I had the circuits set up again. While Boo loves me, I still had to tear down the tests for dinner. So, things needed to be set up again. This time, I included a small glass cup and a meat thermometer. Into the cup I set the sensor, wrapped with a grounding wire with a sender wire on the end. With the circuits powered by a big auto battery, I poured boiling water into the cup. The gauge moved from pointing down to 9 o'clock (correct for water boiling-point) which should be around 210*F. I thought I could confirm the test with the meat thermometer, but instead I found that the meat thermometer was not correct. How fun! In subsequent tests, I discovered that the probe on the meat thermometer needed far more surface area to reach the correct temperature reading. So, rather than triangulate on the proof that 9 o'clock means 210*F, I learned that my meat thermometer reads cool unless at least an inch of it is inside the target. Good to know for future baking. Anyway, this test passed.

Pressure

decoding the no-number gauge

Testing the pressure was not as fulfilling as testing temperature. For temp, we know the boiling point of water so if the gauge doesn't rise to halfway, then we would know that the sensor is on a different resistance gradient than the gauge. That would have been a real bummer. For pressure, I needed to depend on my Mity-Vac. The Mity-Vac has already proven its worth of the years as I have tried to bleed brakes and pull fuel through lines after running the tank empty (I know; bad owner). It also has a pressure setting though. With the circuits wired up, I set to it. I keep a short stretch of fuel line in my Mity-Vac kit, and that fuel line is the perfect size for the threads of the pressure sensor. I threaded one end onto the sensor and put a basic auto-adapter from the Mity-Vac kit in the other end. To that, I connected one of the short clear lines and then the Mity-Vac. This set up allowed me to send a pressure signal to the sensor. I jammed the grounding wire between the end of the fuel line and the sensor housing on the threads (establishing a viable ground) using that friction to hold the wire in place.

around 15 psi and around 1 BAR

Once everything was powered, I simply applied pressure with the Mity-Vac and watched the needle rise off the post. Translating "Druck Pressure" to PSI is a brief exercise in understanding German. Druck is effectively barometric or BAR and 1 bar = 14.7 pounds per square inch. So, to confirm the gauge is operating correctly, I needed to get "Druck" = 1 when I had just shy of 15 pound of pressure on the Mity-Vac. That succeeded: the needle on the gauge climbed to 1 and then slowly dropped as the pressure was slowly released, matching the needle drop on the Mity-Vac. I would have liked to have proven the mid-range of the gauge, but the Mity-Vac and my cobbled-together plumbing couldn't handle more than 15 pounds of pressure. I thought about getting my air compressor involved, but concluded that the test was already successful.

Conclusions
So, what did this tell us? That sensors bought today will correctly align with a gauge pulled from an old Porsche. We demonstrated earlier that the gauge would fit in the spare or "third" hole. I think we have a winner and I'm beside myself with excitement about it. I will clean everything up, replace some bulbs, run a wire bundle, and road-test everything, etc. when I have some time outside with Hapy. You can see in the pictures on this post taken during the tests that the gauge now sits nicely centered in the hole. I think it looks fantastic, but it's my bus, so its my choice. Knowing that I will have visibility into the lubrication system, and can start analyzing whether the oil temps are leading the temperature increases on long climbs is pretty exciting.. for me anyway. If my theory is proven, I can start looking into improving oil temperature management so climbing hills or mountains eventually have a much reduced impact on engine temperature. First, we need to see the data.

Thanks, as always, for following along. More next time-

Oil Pressure and Temperature (Part 2)

In my last post about adding gauges, I suggested 6 possible alternatives. Today, I focus on the viability of option 5: fitting a Porsche combination gauge into the spare hole in the original bus dash pod. I do want to stress that I bought pretty much everything for this project before the end of January, before we knew what this CoVid-19 thing was about. The only exception is the oil temperature sensor. I had purchased the wrong one (English threads not metric) long ago so I needed to get one that would actually fit. On with the post...

Porsche Combo Gauge

blank held up by napkin-holder

First, we start with the question I've been dying to find the answer to: will a combination gauge from a Porsche fit well enough into the third hole on the bus dash to provide oil temp and pressure information? The short answer, I think, is "yes". I'll pause for a second to let that sink in. Yes, that gauge will fit. But, will it look good and do we need to do anything to the dash or the gauge or both to make it fit? Ay, there's the rub. In this case, quite literally... there's a rub. As in, the left-most edge of the combo gauge rubs against the large plastic column in which lies the speedometer. Could it get wedged in good enough? Sort of, but it would look pretty janky. So, could we shave off some of the outer edge of the combo-gauge to make it fit? Maybe. Alternatively, we cut a small section out of the plastic column, and have the combo gauge nestle into the small section. Either way, something needs to be damaged to further the investigation.

almost exactly 4" diameter

the original blank 3-5/8"

So, what's to do? Well, it turns out, my dash-pod, dash board or whatever you want to call the plastic thing that is held on with 4 screws, is cracked. In fact, there is a break running from the bottom edge of one of the climate control sliders all the way through the bottom of the dash-pod. And, the upper left corner is broken off. Cosmetically, it really should be replaced, but it does hold to the dash, and hold the gauges well enough. When I contrast it against the nearly flawless Porsche combo gauge, it seems rather obvious to me which one should be sacrificed to learn how to make it fit... or if it can fit: the cracked and flawed dash.

Before I start pulling out the dash, it would be wise to take a bunch of measurements. I took pictures of them so others can see how close everything is before I touch anything. This includes a test fitting in the cab of the bus. The pictures are strewn across this posting.

Pull the DashPod

combo gauge with glass
from blank on top of it

The dashpod is held on with 4 long Phillips screws. Remove them and stow them safely. Then, reach behind the dash and un-thread the speedometer cable from the speedometer. If you have the plastic tips on your climate control levers, remove them. Now, you can lift and tilt (top downwards) the dashpod so you can get to the wires behind. I can't stress this enough: do NOT just start pulling wires. Identify each one and label it with a toe-tag, a piece of folded over tape, whatever it takes so you can correctly put it back in at the other end of this exercise. Move slowly, be patient. The slower you go now, the more you will appreciate that you took the time at the other end. Once all of the wires have been individually labelled and removed from the back of the dashpod, you can remove it all the way by sliding it towards the center of the bus.

Measure, Check, Measure, Check then Cut
Even though this dashpod is cracked and will be replaced, I have no idea when I could get a replacement. Worldwide commerce is slowing to a crawl, my favorite parts places have been sending notices that they are shutting doors until further notice, etc. So, even though I expect to replace this dash, it could be months before I find a viable replacement. So, I will treat this exercise as if this is the only dashpod in earth.

test fit cockpit side

test fit from "above"

The picture on the right, here, shows where the interference points are. First, at the bottom of the photo you can see that the black ring around the Porsche combination gauge hits the side of the speedometer section before it can fully settle into place. This uneven look can be seen through the rear (front-is-front) of the dashpods in the picture on the left. It may not be obvious, but the oil pressure gauge is slightly less visible than the oil temperature gauge. The two idiot lights should be in the exact left-to-right center of the circle, but they are offset to the right. What cannot be easily seen from any picture is that the edge of the gauge hangs up on the little plastic tang between the threaded hole for the screw and the edge of the gauge hole. I started with those: I cut a shallow channel in-between with a hacksaw.

hole cut in speedo side

trimmed backing

Once I could get the gauge to settle in, I could mark where to cut into the speedometer slot. I covered the area with blue painter tape and marked up ideas with a sharpie. The hole ended up less than an inch from gauge top to gauge bottom and less than 1/2" in depth. It is oriented just below the edge of the outer gauge ring so the combination gauge can slot in. Ultimately, the gauge does not need much space, but it needed a little.

Last, the metal backing plate on the original dashpod needs a trim. It is not as obvious in the picture, but that slight overhang is enough to knock the combination gauge off-center. I marked the short (about an inch long) stretch that needed to be removed along the edge. With tin-snips, I cut that section out and then refined it with a file until the gauges could nestle together. Now, the backing plate threads back on without pressure and the combination gauge sets perfectly centered.

Gauge Wire Identification

labeling wires

If you are thinking of doing this, I strongly suggest that you verify the combination gauge you acquired still works before you grab a saw. I found this kind of fun. The gauge that I was shipped was removed from a late 70's Porsche and then the wires were cut. This is important because I would have had much more research to do had the seller pulled all the wires off. Now, there was a strange green over-spray on most of the wires, so my identification instead was based on what they were plugged into. Similar to the wires removed from the old dash on removal, I labeled these wires as well. There were 2 circuits of particular interest: brown and blue with green dashes. These were the ground and switched 12V respectively. These 2 circuits wired multiple things together, and in the case of the blue/green wire, it had 2 open ends. I think that is to prevent one failing gauge to eliminate the viability of the other. Once these were identified, the signal wires for the various gauges and idiot lights were easy to identify. I removed the gauge lights, after I demonstrated that the ground wires were also grounding those light sockets.

Validating the Gauge

testing oil temp gauge

Now for the fun part: testing the gauges. I grabbed an extra deep cycle battery I had in my garage, and set it on a grub-towel on my kitchen table. My wife loves me. To the positive side of the battery, I connected one blade of a 3-amp bladed fuse. To the negative side, I connected a 16ga blue wire and connected the other side to the ground wire from the gauge. I took a long 16ga yellow test wire which had a female disconnect on one end and bare wire on the other. The disconnect end fit properly onto the open blade on the fuse. I touched the wire to the 12V switched and the gauge clicked, sending the oil pressure gauge to the top. I figured this could be a bad thing, but, undeterred, I removed the wire from the fuse blade and went looking for my senders. I figure this behavior could be a way to signal the driver that there is no longer any signal coming from the oil pressure sender, since I did put the power on the oil-pressure switched 12V side.

combo-to-speedo "front"

I started with the oil temperature, after that odd reaction the oil pressure gauge had to power applied. To simulate installation, I ran another ground wire from the pre-existing blue one around the threads of the sensor. I then ran a jumper from the sensor signal to the temp gauge input. When I applied 12V to the switched circuit, the temperature needle moved off of the bottom post. It wasn't much, but that told me it was reading something. Considering that a success, I did the same thing with the oil pressure sensor. This time, instead of leaping to the top of the gauge, the needle lifted just a hair off the bottom post. Again, I took that to mean success.

I am concluding that the modern VDO sensors will activate the original VDO gauges. I need to conduct further tests, but this post has already gotten really long. I'll post on those tests next time. And, I'll post what the gauge looks like after all the trimming.

Thanks, as always, for following along. I must admit, if you are looking for a can't-leave-home project, this one has been a lot of fun. Please stay safe, and more next time-

Oil Pressure and Temperature (Part 1)

While working on the TDI install retrospectives, it occurred to me that I did not have visibility into the oil system. It bothered me that I did not even have an idiot light to tell me my oil pressure was too low. So, I started working through how I could solve. Today's post starts us on that journey. I started this before the winter holidays, but I'm only really getting down to it now. I had a significant Paulie Axiom event this past week that reminded me of this old post. The Axiom still holds true; I have new examples almost every week.

Sensing

VDO 360 006

It starts with collecting the oil temperature and pressure information. As I mentioned in the final retrospective (See TDI Install Retrospective: ECU Dashpod and Sensors), the pressure sensor in the oil filter housing does not send an "honest" pressure. It somehow sends a relative pressure, so you can't just tie into it for a gauge. I don't want the ECU to freak out because there isn't a pressure signal either, so we need to find a good spot. Our friends on TDIClub (RacerTodd, specifically, on this thread) have located a good spot for collecting the oil pressure and temperature from the oil filter housing. In the picture at the bottom, nicked from that thread, the yellow circle is the spot. The thread size, according to him, is M10x1, which aligns with the thread size/pitch of a standard VDO sensor.

So, with this information, I collected (in late Fall, 2019):
VDO 360 006: oil pressure sensor with the dual idiot light (7psi trigger) and gauge posts,
VDO 323 423: temperature sensor (300*F/150*C Max),
VDO 323 088: temperature sensor (250*F/120*C Max) and
VDO 240 850: sensor splitter / adapter so 2 sensors can run out of one spot

I am concerned that I need an adapter because the temperature probes are so long that it would bump up against something inside the oil filter housing. We'll see. Just in case, I had a sensor that was 1/8npt thread so maybe I can figure out an adapter instead. I also got a shorter sensor that tops-out at 250*F/120*C. I'll get to the sensor stuff later.

Wiring

note unused spot to right of speedo
AND a gauge where your right
knee would be. Huh?

I had intended to run one thick cable from stem to stern for all of my not-typical-bus wiring. Then I considered each circuit individually. This led me to a different decision: 2 thinner cables. Here's why: In the ECU Dashpod, etc posting I linked above, I listed out a number of circuits. Many of these wires run from behind the dashboard (or at the foot pedal) to the spare tire well at the very rear on the driver side. When I add in the desire for idiot lights and a couple of gauges, the number of circuits grew by 5 or 6, but none of them went to the spare tire well. They all ended in the engine compartment where the sensors were. So, rather than plan for a thick cable of, say, 20 wires, I can split that effort into 2. First I will focus on the 6-wire cable that will deliver the signals for the oil temp gauge, the oil pressure gauge and the oil pressure idiot light. Eventually, it will also deliver an Alt idiot light and a low-water idiot light. I'll leave the last wire open so I have it available for something I haven't thought of yet. I'll post on the snaking of that cable after it's done. Depending on the weather, maybe that'll happen soon... or in mid-July. Love those wet Oregon springs.

Gauge(s)
Last consideration is the gauge or gauges. With VDO sensors, VDO gauges are the logical choice. Buses delivered with VDO stuff, so this is staying consistent with original intent as well. Most bay window buses have a blank spot on the dash to the right of the speedometer. Some folks have retrofit a clock or a tachometer in there. Mine is the standard blank plate held on with 2 8mm screws from behind. The visible space within this hole is about 80mm in diameter. A typical VDO gauge is 2-1/16" (52mm). Simple math says that 2 52mm diameter gauges cannot fit into a single 80mm hole. So, I thought about how I could get creative.

which one is the altimeter?

Creative idea 1 - have the lower half of one of the gauges hidden under the bottom of the ring. This sounds great in theory, but the 8mm screws which hold things to the back of the dash are located at 12 o'clock and 6 o'clock. The 6 o'clock one would prevent any gauge from hanging below like that. I considered setting it off to one side that way, but in order for the housing of the gauge to clear the screw, it would be oriented at 4:30 or 7:30 and look real janky. Plus, the upper gauge would be pushed so far up, the top edge of that gauge would not be visible. So... idea FAIL

Creative idea 2 - get one of those 2-gauge housing things and set it on top of the dash. Okay, this isn't terribly creative. Quite the contrary, this is the obvious idea, but to me it looks so bad. Just look at the picture above. I mean no offense, but the aesthetic just isn't there for me. Currently, my dash top is clear. If I were to add one of those, it would meet the need, but I think the looks would bug me every time I got behind the wheel. Beauty is in the eye of the beholder...

classic 2 gauge housing

Creative idea 3 - get one of those 2-gauge housing things and attach it to the lower front of the dash facing up. To give you a better idea, consider that when sitting in the driver seat if you were to look down at the steering wheel and look through it, you would be looking at a warning sticker attached to the steering wheel support. If a gauge housing were placed there, it would be relatively easy to mount, easy to see and it wouldn't get in the way of driving nor your legs, unlike the first dash picture where the gauge is right where your right knee would be. This was the leading candidate until I had creative ideas 5 and 6.

Creative idea 4 - get one of those 2-gauge housing things and attach it to the shelf up where everyone else has sun visors. The problem with this one is that the space between the shelf and the ceiling is too short to fit a housing, so it would have to be attached with just a simple ring-mount. They are not pretty, but would work. I would also have to route the signal wires plus the illumination power up there from somewhere. The extra wiring would prove difficult to hide, further detracting from the aesthetic.

part number 911.641.103.x

Creative idea 5 - get a Porsche combo gauge and fit that into the spare hole. "What's that," you ask. Porsche has VDO gauges stock just like the bus does. They have a 100mm diameter dual gauge that has the oil temp and pressure as well as integrated idiot lights for the alt/generator, and oil pressure. Some of them also included a brake warning light. These run for over $125US on eBay. The biggest problem with this idea is the overall size. If the gauge is 100mm but the hole is only 80mm, then we lose 0.4" in radius, which may effectively block the visibility of the outer edges of the gauges, just like in creative idea 1. There exists an 80mm version of this combo gauge, but I have been unable to find it used, and new it is, like, $325US (well, it is a Porsche part), so that's not going to happen. I don't think this size was produced for many models nor years.

Creative idea 6 - get the participant bits of a Porsche combo gauge and make your own. This sounds a little nutty, but hear me out. If all I need are the 2 small gauge bits, and I can eliminate the extra space between and around the 2 gauges, perhaps I can assemble a combo gauge that fits. I don't know about idiot lights, but first things first, can just the 2 gauge pieces fit? I mean, somehow the Porsche and VDO people figured out how to make one fit into 80mm, and I sincerely doubt they made special one-use gauge innards to fit that slightly smaller footprint. Considering what this guy did, my idea is not that far-fetched. He built his own multi-gauge from the working innards of off-the-shelf VDO gauges. Very cool.

---

nicked from TDIClub

If you've read this blog for any length of time, you know I have to try ideas 5 or 6. Could it be more expensive and ultimately look worse than any of the other options? Absolutely. Could it possibly not even work, forcing me to do one of the other options? Most definitely. But it is this kind of thinking that got the TDI engine into the bus in the first place. So many people said that couldn't be done, or it was a bad idea or that the engine couldn't withstand running at whatever RPM or putting a water-cooled engine into an air-cooled bus was blasphemy. I was even told to sell my bus to someone who would appreciate it and go buy a mini-van or a Sprinter. blah, blah, blah.

So, while we were just learning about some new virus discovery in China (not a dog whistle, just saying that as a reference to a point in time) and a possible first case in the US, I bought one of these Porsche combination gauges off of the UK eBay site for less than $125US delivered. I will see how it looks in the blank spot before I start removing the individual gauges and update as I go. Unfortunately, the VDO folks don't know much about their old no-longer-available gauge components like these, especially when they were built specifically for a car manufacturer. I direct messaged with their staff, and that was fruitless. So, while the old Porsche/VDO combo gauge may look the best, supporting it long-term will be difficult. Ultimately, I may either have to go with one of the other options or build my own out of widely-available VDO gauges that I tear apart like that 914world poster did. I have both an oil temp and an oil pressure from VDO sitting in my garage from back when I had the original engine, so going that direction would kind of be free. Just time consuming. Here's a PDF describing a few gauge-cracking methods if you want to give it a try yourself.

Thanks for following along, and if you have any suggestions for other ideas, or opinions on the ideas I've floated, I am absolutely open to hearing them. I just have to try idea 5 first.

Dashing Thoughts

Today's post is just musings about how to improve my visibility into the health / well being of the diesel engine pushing the VW bus without retrofitting a NewBeetle or MK-IV Jetta instrument panel into the spot where the current original dash lives. This is actually all related, though it may seem a little scattered. I blame the holiday season.

Speed
With a planned change in tire size, the speedometer will no longer be accurate. It's really more of a guide as it is. I know there are converter bits that can be attached to the cable to gear up and down, but I'm not really sure how well those work. There's the option of doing surgery on the speedo, like this page here suggests, but that's a little scary. Still, if I want the original speedo to show the right speed after putting larger tires on, is does read like a viable alternative.

I could figure out a means of getting the speed from a Hall-effect sender. These are basically a magnet on the rotating wheel and a magnet sensor picking up the magnetic field as it passes. This signal is effectively a square wave that an electronic speedo can interpret. OR, the computer could know how fast I'm going, if I route the signal to the ECU. But, I'd still see the old needle on the original dash, showing the wrong speed. I like the idea of being able to keep the original dash operational so I may need to do something to the original speedo no matter what I do about the ECU.

Fuel
I am currently unable to determine how much fuel is in the tank. I tried swapping out the fuel gauge when I thought I had confirmed the sender was good, but that didn't work. So, I have to consider that the fuel sender has failed. If replaced with a ALH-ranging sender, the computer would know the fuel level. But, that would also mean that the fuel gauge on the original instrument panel wouldn't work. So, maybe I could figure out a way of installing an original sensor in the tank and then splicing a second signal from that, but shift the signal in the second signal to match the modern sensor.
Original VW Bus sensor       TDI
10ohms (full)                          35ohms
75ohms (empty)                    285ohms

If the bus range is 65ohms and the TDI computer expects a 250ohm range. For each ohm change in the sender, I'd need .26 ohms of change. Plus, the floor resistance would need to be increased by 25ohms. I haven't done electrical work like this since high school, but I'm not sure how this would work. In the table below, I've split the target ohm values into 13 5ohm increments from the original bus side. The diff column represents the difference between what the sender would provide and what the TDI gauge would expect. The step increase is the additional amount of resistance needed from the more-full to less-full increment. Nothing is simple.

	Bus		diff		step increase		TDI
full	10		25		25		35
	15		39.23077		14.23077		54.23077
	20		53.46154		39.23077		73.46154
	25		67.69231		28.46154		92.69231
	30		81.92308		53.46154		111.9231
	35		96.15385		42.69231		131.1538
	40		110.3846		67.69231		150.3846
	45		124.6154		56.92308		169.6154
	50		138.8462		81.92308		188.8462
	55		153.0769		71.15385		208.0769
	60		167.3077		96.15385		227.3077
	65		181.5385		85.38462		246.5385
	70		195.7692		110.3846		265.7692
empty	75		210				285

Maybe, I could get a converter like the Fuel Gauge Wizard. These are designed to meet this problem for any gauge/sender pair which provides less resistance the fuller the tank and is empty at 500ohms or less. I could splice it in to the original wires so the stock signal is untouched while the new signal goes to the ECU.... hmm... Then, the stock gauge could still read while also informing the ECU to support the digital gauge.

Back to Dashing
If I had solved the fuel and the speed, only the turn signals would remain from the original dash that I'd need to retain. I wonder if there's a way of telling the computer that the turn signal is on....

Continuing down this mental thread, the space available for dash concepts: 13 or 14" across. 5" high. So, fitting a Jetta IV instrument panel (even if I wanted to) wouldn't fit. The one that came from the donor Beetle definitely wouldn't fit. Maybe a tablet could. A typical 7" Android is 19.2cm x 12cm -or- 7.559055" x 4.72441"

Perhaps, if oriented such that the thick "bottom" (or the far right end in the picture here) were set on the outer edges, we could support two screens for a dash.

Seeing that there really isn't an iOS comparable, I looked around for a cheap tablet. $40 gets you a bluetooth enabled, 4GB tablet (link). I couldn't use two of them without modifications to either the vent / heat controls or something else more drastic. Still, its an interesting mental exercise. Maybe it's worth only doing one, covering the blank spot where a tachometer should have been installed stock from the factory (Really VW?, Really?) and most of the speedometer, leaving the cluster with the fuel gauge, turn signals and idiot lights still visible. If the tablet can be easily removed and installed, I could pull the tablet out of the way at will, leaving the stock dash in place. I got one of these tablets just for laughs and the low price comes from the weak battery. Still, this is an interesting idea.

So, assuming the tablet isn't unattractive, how to get the engine computer to tell the tablet what's up? There are a surprising number of tools out there for this, actually. I went and bought this one from scantool. It came with free software for the tablet which I played around with a little bit.

For now, this stuff is sitting in a heap while I consider how I want to address the fuel tank level sender. Ultimately, the sender needs to be replaced, and it makes the most sense to just replace it with an original ohm-range sender. If I want to go further with the ECU stuff, I can do the fuel gauge wizard, build a hall-effect speed sensor and really jump into the electronic dash all while keeping the original functionality.

That's it for this week. Thanks, as always, for following along. With the holidays upon us, and family descending upon us, I may not have much time to post. I am taking some time off, so I'll have some time to generate content of course. I just may not get to telling the stories until some time in January. I appreciate your following, rare comments and more regular emailed thoughts and ideas. Please keep sharing.
Last, if you're in the Portland area and need a place to fix your daily, I'd be hapy to help. You can even use my driveway. Hapy Holidays and Hapy New Year-