How works a vw thermostat (long read, data heavy)

[volkaholic1]

I am the guy that makes brand new, german style thermostats. I spent 2 years doing research, then 2 years making rebuilts, before going 100% new in 2012. I sell all the temp ranges, 75-80 for the 36hp, 65-70 for the carb bug, 80-85 for the FI bug, 85-90 for the type IV, and even 23-34 for the porsche 356. Folks are always asking me questions like "my oil temp is 180F, are my flaps open?" I drove 6 miles to the store, are my flaps open? Its cold here, are my flaps open" etc,,, All I could tell folks was that if you have a 65-70C stat, it will begin to open when it gets to that temperature range. I really wanted to completely understand how these things work in the real world. So I build a setup to measure it. I am using my dad's 74 bug as a baseline, it is almost totally stock. when he got it, it was missing the flaps (of course) the hoover bit (of course) and even the lower cylinder head tins and cylinder head deflector plates. I added all those back and powdercoated it orange and silver. It is a 1600 dual port, the only non-stock parts are aftermarket muffler, bosche 009 with pertronix ignitor, someone cut the pre-heat tubes off the intake manifold, and you can see a mount by the carb for an exhaust gas recirculation solenoid that someone took off.

I decided i needed to measure a few things, so i added some gauges.

going L to R, i have oil pressure, oil temp C, the 3rd gauges is an intercooler air temp gauge, but the top is measuring the engine compartment temp, the bottom measure the hot air that is blowing over the thermostat, the readings are in deg F

the last gauge is a high accuracy voltmeter. I use this to measure the actual postiion of the thermostat in operation. It took me 4 prototypes over several years to develop, with many missteps in the process.

prototype 1 uses a radial potentometer, with a curved gear welded to the passenger side flap, this gear turns a small gear on the pot, alignment was a total pain, i just could not get it right, no matter how many times i welded, rewelded and beat on it, the gear would not stay lined up 100% of the time, so would either miss teeth or bind up - but it does look really cool!

protoype 2 is the same pot with gear, but mounted to a rack at the stat itself, now when the stat expands, it can wiggle a little bit - alignment issues again,

prototype 3, I switched to a linear resistor, this is a switch like you might see mounted in a graphic equalizer (do you remember those?) , this actually worked, but the switch is not made for high temps and it melted

Prototype 4, is an actual industrial linear potentiometer, 13mm stroke, rated for 300F, i finally got something that works

So i started taking short trips 5-10 miles, to see what happens, I record the readings every mile for short trips.

Findings, the thermostat starts to crack open within a few minutes of getting going. Well below the 65C temp rating.

I was shocked at how hot the air in the engine compartment is. The sensor is mounted on the back wall, only a few inches from the fan inlet, so surely this is the air that is used to cool the engine.

Started from ambient temps, which for me now are running 70-85F (22-30C), the air in the engine compartment rapidly increases. The highest temp i have seen so far is 66C -151F, the overall average for all my driving so far is 46C=115F, not exactly cool air for the engine at all.

As far as how hot is the air the stat is seeing, well obviously it must get to at least 65C, or the stat would not work.

It really only takes a few minutes to get to that temp, I had to take longer trips to get it to get to steady state condition, but based on my limited data so far, it looks like a 65-70C stat holds the heated air from the engine in the 100C range - 212F. This does not seem like an unreasonable number to me.

this i think is important to understand - steady state operation

I finally drove 50 miles to work, and 50 miles home, I did all back country roads, trying to stay around 50 mph, i did have a few stop signs and lights and some 35 mph in town driving, but tried to be consistant. When the engine gets to steady state, the stat cycles between 85% & 95% open, rarely have i seen it going full open for more than a minute or two. This chart shows sensor voltage (bottom) vs mm of stat displacement (left) vs stat % open (right), the stat spends most of its time constantly openign and closing small amounts to keep the engine temp constant. Think about it, that makes sense, it should do that.

I have been waiting years to do this last test, again on this drive to and from work, gong to work, i kept everything normal, but coming home, i tried the "tennis ball trick" - you know, folks put a tennis ball between the decklid and the latch to open the decklid slightly to get more cooling air in. Well,,, IT WORKS! This chart compared miles driven (bottom) with deg C (left) with stat voltage (right). This stat readings are dashed lines, note that they are almost identical, the stat pops open within 10 miles, then slowly closes and starts to cycle open and closed slightly. You can see that engine compartment air temp is like 35C - 95F LOWER with the tennis ball, stat temp lags behind about 10C - 50F with the tennis ball, and oil temp lags behind with the tennis ball but looks to be coming together if i drove farther.

I think that is some pretty amazing stuff to learn.
Other things i would like to check
- what if I run the 75-80C stat?
- what if i run the 80-85C stat?
- how about cold weather driving?
- what if we installed this on the type iv engine?
- any other ideas on things to check?
_________________
Volkaholic

[Max Welton]

Very nice!

The air going into the fan (your firewall-mounted sensor) is a lot hotter than I thought it would be. I'm starting to really appreciate the type-3 arrangement that more effectively presents cool outside air to the fan.

Max
_________________
1967 Type-3 Fastback
Under the Knife https://shoptalkforums.com/viewtopic.php?f=24&t=151582
Home Stretch https://www.thesamba.com/vw/forum/viewtopic.php?t=767836

[AlmostHeavenWV_VW]

This.
Is.
AWESOME


Well done


Love the temp charts and readings.


Neat to see the actual data regarding the 'tennis ball' trick!!!
_________________
1973 Standard Beetle
1600DP AK case
Solex 34PICT3 Carb
Bosch DVDA 205AJ Distributor

[theDrew]

I love this. well done
_________________
Turbo 2276 MS3X build http://www.thesamba.com/vw/forum/viewtopic.php?t=673125
1985 Vanagon Campmobile w/ 2005 EJ25

[Steve Arndt]

Thank for doing this test.
_________________
Steve
Steve's 87 Syncro project

[Emeraldlion]

I honestly would not have expected the tennis ball to make THAT big of a difference. Would be interested in seeing decklid standoffs, and what the temp run on a closed engine bay like a 67-earlier without vents

[raygreenwood]

The engine compartment on totally enclosed set ups....meaning anything but type 3 and bay bus.....gets damn hot.

Which is why......type 3 and type 4 cars pull their engine inlet air from outside in a dedicated duct.

I have measured average temps in my type 4 (412 and 411s) at a constant 185+ on long highway hauls in hot weather at 70-ish. In and around town....it's about 125 F.
But its a fully enclosed engine compartment.

The inlet air is always ambient outside air.

Very nice work!
Ray

[dawie]

Nice.

Any chance of replacing that 009 distributor with an svda?

[Max Welton]

I can add some type-3 configuration data. I am running a Megasquirt EFI system so I have lots of data logging available.

I also have a sensor located on the right sled right behind the thermostat. I am using it as my CLT to adjust enrichment, sort of a choke function for efi. The temp sensors I use look to be the same open-element ones you are using.

I do not have anything to measure air temp going into the cooling system, but it's a type-3 so I would expect it to be close to ambient.

My Intake Air Temp is placed in the type-3 TB.

This particular log was taken in 2016 with ambient temp ~90ºF. Total time was around 20 minutes with the air temp down by the thermostat topping out around 190ºF. I have some longer logs that more clearly show regulation of the temp in that airstream.

Max
_________________
1967 Type-3 Fastback
Under the Knife https://shoptalkforums.com/viewtopic.php?f=24&t=151582
Home Stretch https://www.thesamba.com/vw/forum/viewtopic.php?t=767836

[RWK]

I had a oil temp gauge in my stock 74 std. bug back in the 80s, on the highway in summer I would leave the deck lid open to the first position and it would stay,( not all the way open) it would stay kind of horizontal, people would always try to flag me down pointing to my deck lid, I would see a 20 degree plus reduction in oil temp, never though of a tennis ball though, didnt want the hood stand offs cause I lived in a cold climate, I use the tennis ball now in my 63 with a hot 1600, deck lid different in the older models and no extra vents.
_________________
73 Type 181
63 Type 113
63 Type 261- 428 071
62 Type 241-378 025 178 530

[sled]

awesome testing volkaholic! thank you.


now..if we could figure out a way to get more air in like the tennis ball does, without it looking stupid like a deckled propped open
_________________
drive your split.

[Emeraldlion]

With the tennis ball or a "platformed" stock latch, I can't really tell that the decklid looks propped that much. Don't think it looks bad. The only other option besides vert decklid ( not realistic for my 67) is 4 or 5 small holes behind license plate. I like the Cal look of standoffs but don't like engine being open to elements that much

[74 Thing]

[volkaholic1]

i do want to try the different stats, and i also make venturi rings for the fan inlet, so i want to give that a try, and i think deck lid stand offs is worth testing
i have never worked on a type iii, so i did not now the air intake difference, sounds like a good idea - i remember in the 80's some folks had a fiberglass "scoop" that ran along the entire vents, does that still exist? anyone ever tried it?
_________________
Volkaholic

[raygreenwood]

[mcdonaldneal]

Great information and a very interesting thread. Many thanks!
_________________
1978 marino yellow Bay Dormobile camper
1969 signal orange Karmann Ghia convertible
1976 martini olive Bay Dormobile camper

Stop dead photo links! Post photos to The Samba Gallery!

[Max Welton]

One problem with the type-3 is that the engine room is like an upside-down tub. When you stop in traffic that tub fills with hot air. That's why heat-soak is such a problem with type-3s that have been converted to carbs.

Max
_________________
1967 Type-3 Fastback
Under the Knife https://shoptalkforums.com/viewtopic.php?f=24&t=151582
Home Stretch https://www.thesamba.com/vw/forum/viewtopic.php?t=767836

[gt1953]

{[One problem with the type-3 is that the engine room is like an upside-down tub. When you stop in traffic that tub fills with hot air. That's why heat-soak is such a problem with type-3s that have been converted to carbs.]}

But the engine is still getting cooler air to cool it right. The when the light turns green cooling in the engine room cools also right.

I do like the non tennis ball tennis ball temps Obviously it works.
_________________
Volkswagen: We tune what we drive.
Numbers Matching VW's are getting harder to find. Source out the most Stock vehicle and keep that way. You will be glad you did.

72 type 1
72 Squareback
({59 Euro bug, 62, 63, 67, 68, 69, 73 type ones 68 & 69 type two, 68 Ghia all sold})

[Max Welton]

[timvw7476]