Flexible Service Interval Retrofit?

Subscribed! Would love to have the flexible service interval (and more importantly oil level) working on my FIS cluster while keeping it coded to US spec for MPG and what not. I read this post a while back when the mod (oil pan and sensor) wasn't as affordable as it is now. I was all ready to do it, but I didn't want to switch to miles per imperial gallon.

The late dumps always have 0 in last digit due to using the CANBUS for the speed impulse.

Nihilator posted something before when he was recalibrating his speedo to display exact speed correctly - you have to manually change the figure in the dump itself, not with VAGCOM I believe.

That's what I figured.

I think its working like its supposed to now! I just have to buy a oil sensor and pan, wire it up and see what happens. Whats weird is that I coded it for a sensor and once in a while it will display the "oil sensor workshop" and flash the yellow oil light. It only happens sometimes though. Most of the time the cluster doesn't say anything about the missing oil sensor although there is a fault code for it. Its a custom US dump with 3B at address 021D so that will be full proof it works with a oil sensor.

I pulled out the 501MH cluster that I was using in it and I am going to play around with that on the bench. Hopefully we can figure out what exact values turn it on and off. Then we can figure out how to make non-MFA US dime-a-dozen clusters compatible with a oil sensor too!

This is huge if it works out.

graeme86 said:

Great work on that!

The equivalent matching sequence address on a 501MH cluster should be at 0x1EB

Values look to be different though to yours quoted.

Both my R32 MFA and FIS clusters (euro) currently have 2B (they have the capability of Flexible Service but are currently softcoded in VAGCOM for Fixed Service Interval)

I will check if changing softcoding also changes this value as well.

US 503MH software R32 cluster has 0C value in your 021D position

So maybe only one particular bit actually activates this feature

e.g I look for a pattern in binary values:

08 = 00001000
0C = 00001100

2B = 00101011
3B = 00111011

Possibly just one (or a combination of) these positions: xx1xxx11
is the critical value that actually switches on the feature.(???)

Click to expand...

Time for some much needed updates. Been busy lately and haven't had much time to devote to this project.

So far all the US immo3 501MH dumps I checked have 08 at address 01EB.

The euro version has 2B at that address.

When I put 08 in my US 501MH MFA, I cannot code for flexible service intervals as it disables it. I tried 3B and it worked just fine as well. I think like gaeme86 is saying, there may be several valid values that will work.

I checked a non-MFA US immo3 cluster (part 1J0-920-906L) and it had 08 at that address.

Problem is that I do not have a regular black immo3 cluster on hand at the moment. So if somebody wants to contribute to this project ,and has the stock immo3 cluster in their car, and has vag-tacho, they can change address 01EB from 08 to 2B and see if it allows you to code for flexible service intervals. If it does, see if it throws a fault code for a missing level sensor and if it tries to light up the yellow oil LED with output tests on VCDS.

I took a quick look at a basic US spec immo2 non-MFA cluster and it had 00 at address 01EB. I'll have to experiment more on that one and compare it to a euro dump again.

Big thanks to grame86 again for converting the 503MH address to 501MH. Saved me a LOT of time!

I also have a Euro T5 FIS dump (7H5920870C) and the equivalent position appears to be at 0x24F

The byte value is 0B so one could speculate that the "-B" is the critical value that activates this feature.

If this is the case, it would be interesting to see what the 0-, 2- or 3- does in the first position.

Just tried 0B and it works too! So I guess the bottom line is that to do the flexible sevice mod, you change X8 to XB and X can be whatever the cluster had originally. Not sure what the first digit is supposted to or what it controls, so I would think it would be best to leave it as whatever it currently is.(?)

Just took a quick look at a immo2 dump and it has 00 at 01EB on a US spec cluster and a euro dump has 33 at that address. Interesting....more should be coming on that.

The immo 2 address might be different - don't they have a smaller overall dump size?

Hmm...good point. I believe they are both 2KB dumps, but my vag-tacho only reads 512bytes on immo2 for some reason.

I'll have to look into that. It might be in a different spot on immo2

WIN!!!




Ok, so its kinda hard to see. The oil light is lighting up in a amber color due to both the red and yellow portions of the LED lighting when I run output tests in VCDS. Again, a 300ohm resistor did the trick on a immo2 non-MFA standard dime a dozen TDI cluster.(1J0920905L) You can get these things for $50 from VW.

It does throw a level sensor code when I code it for flexible service intervals too! I did cheat a bit though, I loaded a euro dump into it to enable it. Address 01EB is not what controls it on a immo2 cluster. Its somewhere else which I have yet to find. I'll probably find out soon when I get time to play with the dumps.

turbocharged798 said:

Big thanks to graeme86 for hooking me up with a euro passat 503MH dump. I loaded it up into the GLI cluster that I had and the odo didn't work correctly due to the passat odo being different, but when I opened the controller, it was already coded for flexible intervals! Sure enough, there was a fault code for missing oil level sensor. Ran output tests and and it tried to light up the yellow low oil lamp, but nothing happened. So I installed a 400 ohm resistor here. I chose 400ohm because it was the same that controlled the red side of the LED.

Click to expand...

I think it is 400 K Omh resistor.
EDIT:
I made mistake it was 0.4 K Omh so it is 400 Omh

400K is 400,000 ohms. That would be way too much resistance and the LED probably would not work.

Deebeaux said:

Subscribed! Would love to have the flexible service interval (and more importantly oil level) working on my FIS cluster while keeping it coded to US spec for MPG and what not. I read this post a while back when the mod (oil pan and sensor) wasn't as affordable as it is now. I was all ready to do it, but I didn't want to switch to miles per imperial gallon.

Click to expand...

Another approach is to enable UK settings, then change the EEPROM mpg fudge factor so you get back to US mpg. With a few tank mpg measurements you can fine-tune it so your MFA mileage readout will be very accurate.

turbocharged798 said:

400K is 400,000 ohms. That would be way too much resistance and the LED probably would not work.

Click to expand...

You are right it was very late last night and I was so tired, and it was 0.4 k omh.
Sorry my bad.

Well, I have good news and bad news.

Good news is that I just received my oil level sensor that I bought off ebay for cheap(it was used OEM).

Bad news is that non of the us clusters I enabled flexible service can "talk" to it. I hooked it up to the cluster on the bench and it still keeps throwing a missing oil level sensor fault. Boohoo.

Spent an hour or so tracing back the circuits on the board and didn't come up with much. There are some empty pads but its very complex and without a true euro cluster in front of me, almost impossible to find out what is missing.

This rots.

I don't think you will be able to do it on the bench. It most likely it goes through CAN BUS and it need to see ECU . I know for sure oil pressure light goes through CAN BUS so it is probably same with oil level light.

Did you wire it up like this( page 12):
http://forums.tdiclub.com/showpost.php?p=367007&postcount=18

Worth a read (relates to Audis but has some good info on the system):

http://www.alm.hk/tiny/tinytog.php

Did you try other different "valid" coding combinations as well (0B, 2B, 3B etc)?

So far I have been working on my immo2 1J0920905L cluster with a euro dump in it. Just incase I fry it for some reason, its not an expensive cluster to loose. I did not figure out what values control the flexible service on this cluster yet.

Good news though, think I got it working finally. It was not easy.

First off, pin 18 is what goes directly to the level sensor. Well, tracing that circuit back revealed a whole load of empty pads and it basically did not connect to anything. So I took a stab in the dark and jumped a pad with some solder. Did not work. Jumped another pad with solder(I was grasping at straws at this point) still did not work, but I noticed it was now throwing a code for short to ground rather than short to open/B+ like it was doing before. So I went back and undid the first jumper. Plugged it back in, and went to measuring block 3. BINGO! 21*C and it said "level OK". No more fault code.

What was weird is that I left the sensor on my bench for over an hour pugged in and it still said "level OK" when it clearly was not OK. Not sure what is going on with that. I sure hope it works better than that once its installed in the car.

I think there may be additional jumpers/resistors needed to allow the cluster to sense when the hood is opened and reset the light. Too bad I couldn't get it to trip on the bench so I can test that.

I am thinking at this point that it might not be a viable solution for most people as it is quickly getting complex to make a US cluster compatible with flexible service. I doubt 99% of people out there would do these mods at this level.

jetta said:

I don't think you will be able to do it on the bench. It most likely it goes through CAN BUS and it need to see ECU . I know for sure oil pressure light goes through CAN BUS so it is probably same with oil level light.

Did you wire it up like this( page 12):
http://forums.tdiclub.com/showpost.php?p=367007&postcount=18

Click to expand...

I know for a fact that it does not go over can bus. It reads the data directly from the oil level sensor from pin 18 green.

This post (with pics!) shows where the Bonnet LED Resistor goes:

http://www.gti-tdi.de/board2/index.php?page=Thread&threadID=3219

Graeme, thanks for that link! I totally forgot about that. Best part is that it showed me a euro cluster and where the missing resistors were. So I put them in along with a transistor and it appears that my MFA cluster is working too!

The thing is that I cannot get the low oil sensor to trip for anything. I want to see if it can detect if the hood opens but I need the low oil light to come on. I have the sensor sitting on my bench, upright connected to the cluster. Measuring block 3 in VCDS says "oil OK" and "21.0*C". Why can't it figure out that there is no oil at the sensor? My engine would have been long burnt up by now if I had to rely on it.

The only thing I can think of is that it uses it along with another sensor to detect low oil.

What sensor part number do you have?

1J0 907 660 B should work for Mk IV okay

The "C" version is for later CANBUS (like Mk V etc) so may not work with your experiment.

Its the B version. I double checked the part numbers before ordering it.

Really cool to read about these efforts, hopefully you'll get it working!

My car has a FIS cluster and a pan with the oil sensor but I switched off the service interval (I keep track of maintenance intervals on a spreadsheet anyways). The thing that makes me nervous about the flexible service interval is that there seems to be so little information about how it does its calculation - presumably there is some formula that VW engineers worked out based on oil temperature, etc., but we don't know what it is (or at least I haven't seen it detailed on the forum). If you do any performance mods to the car, it would seem this formula might well become untrustworthy - and in the case of something like this where a cluster is not originally intended to process the data, who knows whether it's doing the calculation right, or at all?

Has anybody sorted out exactly what data this sensor logs, and where it gets stored over time (somewhere in the cluster EEPROM)? Seems like monitoring this directly could be more useful (and fun in that nerdy cluster-mod way) than just seeing the "change oil now" light, especially if occasionally compared against UOA data.

I haven't read this whole thread since it's almost 10 years old, but does anyone know if this is possible with a MkV?

For the life of me, I still can't figure out why the cluster thinks the oil level is fine when its on the bench when its clearly not in any oil. It should trip it right away and turn on the low oil light.

Maybe I got a bad sensor?

I think at this point, most people would not be able to convert a US spec cluster to flexible service very easily as it requires several parts and not just one resistor. Oh well.

If the sensor is like a thermistor, try putting a pot there instead and see if you can trip it. Otherwise.. Maybe the "low level" signal is getting filtered out somehow (cap, resistor, lack of both or either where there needs to be one).

Also try submerging the sensor in some sort of liquid and then removing it. Maybe it needs to see liquid to know the range.

As I previously posted (Post #276), here is a link that explains how the sensor works:

http://www.vaglinks.com/Docs/VW/Misc/VW_Service_Interval_Extension_SelfStudyGuide_224.pdf

(around page 21)

Thanks graeme, very interesting, should have read deeper! Too bad there's no easy way to display the oil temperature the sensor is reading on the cluster

graeme86 said:

As I previously posted (Post #276), here is a link that explains how the sensor works:

http://www.vaglinks.com/Docs/VW/Misc/VW_Service_Interval_Extension_SelfStudyGuide_224.pdf

(around page 21)

Click to expand...

Thanks for the link, although I still don't see why it would not trip on the bench. I'll have to try dipping the sensor in some oil and hopefully then it would trip.