1.6 FSI Map Controlled Engine Cooling Thermostat Woes

Re measurement blocks - when I was trying to work out my ODB reader's channel contents I found the following link to an excel spreadsheet detailing the list. It isn't perfect for my A2 or my A6 but it is reasonably close to both (ie you have to take into account some channels missing in smaller cars because they are information for cylinders 5-12 and blocks for temperatures on the second bank, both sides of the exhaust and so forth that we simply don't have).

Unless there is something very different to that table, you might also have a squint at the channels in the 130s as these appear to be related to various temperature / coolant circuits. Just noticed also channel 95 that may be pertinent to 1.6FSi - the coolant temperature is included in data blocks relating to the intake manifold change-over - isn't this one of the critical elements in how the FSi works (and goes wrong?) most efficiently - which could have effect on the mpg. Good luck.

nyet.org/cars/info/VCDS%20List%20of%20Engine%20Measuring%20Blocks.xls
 
Last edited:
1.4 is nowhere near as complex. search the ssps, you should find the engine-specific one ( though aua is *very* similar on many levels).

- Bret
 
Thanks Robin(@Robin_Cox), that looks the kind of reference I meant and should be most helpful. I had a quick scan and channel 130 and there abouts looks very relevant. You say 'channel' is this what VCDS calls Group'? The lower inlet manifold and its flaps were refurbished by Tony (RIP) at A2 cars two years ago (only 7000 miles since) so should be fine, certainly no fault codes and the car drives A1, but I must admit I was under the impression flap function was dictated by RPM, yet I can see on a general level coolant temperature might enter the equation.

Thanks Bret(@bretti_kivi), I will run the German thread through Google translate and take from there. I have two friends who speak fluent German but they know nothing about cars and any specialised vocabulary beyond the word 'engine' might be beyond them.

@Catnip64, I have scanned that ssp before, I think I picked it up via a post from Spike, but now I have some data temperatures I will give it a serious read over Xmas. With the data I now have it will interesting to see if it matches diagram 222_019. Thanks.

All I am really hoping for is someone to say those graphs look fine or the opposite.

If any TDI readers are thinking 'why not just replace the thermostat it's only ten quid(ish)' can I point out I have estimated parts+labour at £175 (+another £150 if by now the very brittle thermostat housing breaks), it is an FSI after all and I am a pensioner!

Andy
 
feel free to comment on the German forum in English if you need specific vocab. "ist" = is, "soll" = should be. So the "ist" is measured, "soll" is what the value should be, whether it's for action taken or not.

- Bret
 
Hi Andrew, I think we're saying the same thing, yes : VCDS use of the word "group" corresponds exactly to the "channels" on my ODB2 reader, with Blocks being the 3-4 readings within that "page". Either way, even if I'm getting my words mixed up, the way you've interpreted what I meant is how I intended. I expect you've noticed already that the spreadsheet sometimes has duplications of a "group" with minor differences in the readout blocks - with the small print explaining which version is likely to be applicable in our case (ie engines with 2 banks vs our 1 and so on).

I can't find a VAG SSP booklet for the A2's 1.6FSi variant (there is one for the Touran BAG engine which is an evolution of the BAD and there are certain differences such as a double cooling circuit, head vs block with two thermostats, no G70 meter etc.). However, the contemporary 1.4FSi from the Lupo (ARR) engine has a guide, and it is possible that it and the BAD are similar in terms of working principles.

http://www.volkspage.net/technik/ssp/ssp/SSP_252.pdf

See page 22 where the logic of the different fuelling modes is explained - if the BAD uses something similar coolant temperature is definitely taken into account to engage the FSi mode. The rest of the manual is a good read in terms of explaining the myriad of sensors working together in various systems to control pollution, economy and so on (many of which impact on FSi vs normal activity) - but also have a look at p49, where the switchable thermostat modes are explained - it could explain why you see active and dramatic discrepancies between internal measured temperature and the temp gauge merely indicating "90" to the driver. Best of luck!
 
Vorsprung durch Technik said:
Hi Andy, how did you get on with this?
Click to expand...

Hi Mark,

Xmas got in the way since my last post and early January has been occupied with preparation for plastering my front garage ceiling, which happens on Monday, a precursor for adding a room upstairs in my back double (near) garage which is happening in the Spring (with Building Control at the moment). I should have a little time next week while the plaster dries to get back to the thermostat. I am wondering if the thermostat is not closing properly allowing coolant through the radiator circuit immediately from a cold start which is why it seems to take so long reach temperature and degrade the mpg, though just a guess. To test this idea I plan to monitor G83 and I will post a graph. The mpg has risen slightly to 33mpg (as opposed to the normal 40mpg) and the car pulls fine, it is just costing me a few extra pennies on mpg but I don't do many miles a month so no drastic hurry to fix.

Andy
 
It is so long since I've posted in this thread I thought a quick recap is in order.

Historical steady 40 mpg suddenly starts dropping rapidly. Scan showed thermostat fault code (post#1), cleared and has not returned but mpg continues to fall, down to 28 mpg at one point recently. Also seems to take an age to reach 90 C from cold. Car drives and pulls well.

I have a theory. The thermostat is NOT closing properly, always a little bit open allowing some circulation to radiator, causing long warm up period and poor mpg (since currently only do very short trips - I should really walk!). This post is about testing the theory.

Following Robin's helpful post above on identifying VCDS measuring channels and particularly the few channels starting at 130, I thought I would focus on the radiator outlet temperature, supplied by the G83 sensor.

This is the channel content from Robin's link.

49996


A lot of this content I do not understand, 'Thermostat duty cycle', 'Heater pre-run potentiometer', '(coolant filling), short trip', 'Result', 'Status' ??????? I am not that concerned as it turns out those boxed in red never change. The 'specified' fields I assume are some kind of target value that act as a trigger to invoke some kind of response.

A snapshot of what VCDS shows, this was at the end of a 25 minute test drive, sat idling in Sainsbury's car park, in an attempt to get the coolant up to 110 C, the point when the thermostat is rated to open. After 5 minutes idling I managed to reach 110 C and nothing changed but looking back perhaps I should have persevered as it may be a lower boundary that has to be exceeded.

49998


If I understand this correctly the binary Status field suggests I have a 'control deviation', whatever that is?

I logged the test drive and this is what the start of the log looks like, this log is a mere 1845 lines long, much shorter than previous log files!, I assume because I am logging far more, but just as much data in total.

50006


Looking at long data files is beyond my human capabilities so next step is get Excel to cough up a graph, I am getting quite good at this!

50000


Oh well, theory debunked back to the drawing board. Clearly no hot coolant is routed to the radiator during the warm up from cold.

A few comments on the graph.

Again curious trough at just below 80C on warmup at 9 minutes, as I said before I wonder if this the point when the cabin climate begins to seriously use coolant heat, and a little blip up on the radiator outlet temperature, but probably just coincidence, maybe be just engine compartment warming.

Operating temperature is reached at 18 minutes with the saw tooth high plateau, but curiously the G83 sensor shows no life for a further 5 minutes, despite instances of operating coolant temperature dropping.

I assume the thermostat opens at 22 minutes with the noticeable rise in reported G83 temperatures. I am impressed by the efficiency of the radiator system, dropping coolant temperature from over 100C to under 20C.

What educational fun.

I am a little behind in posting, the test log date stamp was a month ago. There have been developments since, three posts hopefully next week.

Andy
 
Last edited:
You must log in or register to reply here.
Back
Top