6R Diesel Polo to Petrol Fun & Games

[nugentp]

I don't have huge experience on the Mk5 Polo (6R platform) but I still got roped into a project to convert one from its original 1.2 diesel engine to a 'somewhat larger' petrol engine from a Mk1 Audi TT (cough). We sourced a 6R petrol tank and related parts for the conversion.

We have said petrol engine running on a test cradle for now (it was bought for not outrageous money without a chance to hear it running in real time). I tweaked the ECU for 'immo off' operation for now and we have some more tests to do before establishing the full list of mechanical refurbishments needed along the way. I have a Motronic guru lined up to help solve the bigger challenge of integrating the ECU with ESP8.2i family ABS (instead of the MK60 system the TT expects) and any other coding or firmware tweakments so the ECU settles into a new home without lighting up the dashboard like the proverbial xmas tree.

Apparently the Mk1 TT instrument cluster will physically fit in the Polo reasonably well, but I worry about all sorts of BCM related gremlins and fault codes if we do that. The other issue is the central display on the TT cluster is heavily pixelated and unusable - seemingly a design feature of those old displays based on the number of threads you see about failures - so we can't use that one at least.

We can't really retain the cluster from the 6R polo diesel engine either as the RPM range available is not suitable for petrol engines.

So... wondering if a set of clocks from a 6R/6C petrol engine (maybe from GTi variant) is the way to go as it should integrate with the original BCM in a mostly expected way, but does anyone have experience of doing such a cluster swap on 6R/6C. This opens three questions:

1. Do we need 'special software' to transfer VIN information to the donor cluster ?
2. Will component protection be involved before the car's BCM (or ABS etc) will accept a different cluster ?
3. Maybe we need to also swap the 6R Polo BCM from diesel version to petrol version as part of this conversion ?

On the 3rd question I really need to find some 6R/6C Polo VINs so I can cross-check the BCM p/n on PL24 to see if there is any sense of petrol vs diesel versions - showing my lack of experience on that platform.

More info will be openly shared here as the project advances - there are some mechanical challenges around things like driveshafts and whether the exhaust will clear the bulkhead etc but we dunno how much pain will be involved until we get there.

 

[nugentp]

Trying to answer Q3 myself after a bit of VIN chasing on PL24 - seems like the BCM variant is determined primarily by things like start/stop capability and the headlight configuration, so I am happy enough that the existing BCM can integrate with petrol or diesel engine & clocks.

Boy oh boy do those fancy 6R GTi clusters fetch a premium price on auction sites with their extra needles and 180mph range - literally 5X the price of regular 6R petrol clocks which only feature rpm (up to 8K) and 160mph speedo (plenty).

The engine will of course start and run without any 6R components connected (ECU is immo off) - what I am wondering about is the level of fault codes and nagging lights on the 6R petrol cluster without an IMMO conversation it expects to have with a more modern ECU and whether mismatched VINs across the main components is something that MUST be solved or is just nice to have for less fault codes.

 

[nugentp]

Let's say I have special software to transfer EEPROM contents from old diesel cluster to new petrol cluster so the info about VIN, IMMO, keys, SKC is transferred over - does component protection also come into consideration or is the EEPROM going to solve everything by cloning the cluster info ?

As expected, the IMMO4 ops of the cluster cannot override the IMMO3 off behaviour for the VR6 ECU as the engine starts and runs on the cradle just fine. A nice surprise is the cluster displays correct RPM on the 6R cluster so that's something.

There will be other challenges to solve as the 6R clocks don't use a gearbox sender for speed input... nor does it send old school VSS signal to the ECU. That's a problem for another day.

 

[2kaa52]

You shouldnt do this, you have Zero knowledge......

 

[nugentp]

"Zero knowledge" you say - thats quite the claim... I do have limited knowledge on IMMO4 and PQ25 specifics, but I have a particular set of skills that usually takes me to where I need to be one way or another so let's just see how this develops.

 

[nugentp]

Just a quick update in my 'zero knowledge' quest here.

After doing a bit of homework, first finding how to hold the NEC MCU in reset so the EEPROM from the AKL cluster could be read with a cheap device I made some progress. Then I invested in a genuine instance of VAG-Tacho (a very useful tool with excellent support).

After a bit more learning along the way I now have the IMMO4 info from the diesel cluster transferred into the 6R petrol cluster. That should stop any nagging from the ABS, BCM etc about mismatching VIN etc, but it turned out I needed to IMMO-OFF the petrol cluster to ensure it stays out of SAFE mode... and doesn't attempt a conversation with an ECU that only understands IMMO3. I already have IMMO OFF in the ME7.1.1 ECU so everything appears to be going to plan. It fires up and runs great - sending signals to the 6R petrol cluster.

Obviously a good quality aftermarket security system is important for this kind of setup where the different generations of factory IMMO systems cannot play with each other, but thats not a topic for here.

I did hit a glitch trying to program a fresh key for this 6R32 frankenconfig with VAGT but thats not something for here - its either some small print that I haven't figured out yet or finger trouble or an incompatible key.

Solving VSS is the next task to tackle while the engine is running in the test cradle - I am expecting that to be easier than IMMO wars on the assumption that I can direct the G22 signal from the gearbox sender direct (or via some kind of isolated buffer circuit) into the ECU as its sense of VSS but I need to see things on a silly scope first of all so I don't blow anything up. Thankfully I have the ECU breakout box for that - all 121 pins or whatever it is.

Anyway - RPM and CTS info from the VR6 engine is being sent to the 6R petrol cluster without any drama - some convenient reuse of CAN Bus labels I expect between 2004 and 2010 cluster but I haven't tried to sniff the bus yet - could be some odd things and I am not thinking about the (w)hole ABS interworking until we get that deep. I do have a highly respected ME7 guru lined up to help with those most tricky parts of the integration but thats a way off.

Component protection (CP) would have made this a far bigger nose bleed, but that doesn't come into play on 6R machines - seems like thats a thing on 6C from 2015/2016 wish.

 

[nugentp]

So, I have made a simple little circuit to generate VSS from the G22 sensor on the gearbox as the 6R cluster does not generate it - that platform presumably relies on the ABS controller broadcasting vehicle speed based on ABS wheel speed senders.

The VSS gen circuit is working perfectly when I have the VR6 TT ECU running with the 6R cluster in that the signal is exactly as I would expect, but the ECU is not displaying any sense of vehicle speed. I have it on a test cradle with gearbox turning in whatever gear.

Does that mean the ECU expects a CAN-Bus message from the cluster/kombi about what scaling factor to use for the VSS signal before it knows what to do with it... or does the 8N cluster actually send VSS info over CAN Bus as the primary method here and VSS is just some legacy / fallback ?

I think I am going to have to open the quiet-quiet piggy bank here and invest in a CAN Bus sniffer once and for all.

If looking at the pic CH1 (yellow) is the G22 input - note that is pulled up to B+ by the converter so it is well above the schottky threshold and CH2 (blue) is the inverted output which is basically a perfect match for what I observed with the OE 8N cluster when it was generating VSS from the G22 input.

PS - happy to share the circuit with anyone interested - its not exactly complex and very cheap to build !

 

[nugentp]

For complete transparency, attached here is the OE setup but just note it was taken in a different gear (so frequency is different) and the channels are swapped over. The G22 input (yellow) is around 8V amplitude as the 'raw' signal coming off the hall effect speed sensor in the gearbox - and you can see the VSS (blue) signal is pure inverse of that coming from cluster tracking the frequency in real time.

My converter pulls the G22 input up to B+ to ensure it cleanly exceeds the threshold for edge/level detection on the little schottky inverter circuit.

 

[nugentp]

And heck here is the VSS generator circuit for good measure... about as simple a digital circuit as one can imagine !

 

[nugentp]

So... a bit more investigation suggests strongly that the 8N ECU simply ignores the VSS pin as it expects to get VSS info via CAN-Bus messages from the cluster (or possibly the ABS controller). I got some bus sniffing to do !

 

[Dr Sheldon]

You shouldn't do this, you have Zero knowledge......

I would enquire the cost of a Polo Gti on finance over 3 years ! Purely to bring "sanity" into the equation !!

 

[iichel]

who cares about sanity, it's a really cool project!

 

[nugentp]

I would enquire the cost of a Polo Gti on finance over 3 years ! Purely to bring "sanity" into the equation !!

Thats a far too sensible idea... where's the fun / opportunity for education / buying new tools in that ?