GEMS MAF Translator Development

[Aragorn]

Not sure if it fits here or if anyone is particularly interested, but i wanted to document my efforts at creating a replacement for the GEMS MAF. As some of you will know, the GEMS MAF is unavailable anywhere other than Land Rover (if its even still available) and costs a fortune (£1000 last time i looked). Aftermarket units are junk, and the pile of working used parts are dwindling as these models get older. After buying probably 4 or 5 used MAFs at this point, i wanted to find a solution.

My thinking is simple. If we can figure out the curve, we can build a "module" which can translate the values from some other cars MAF, over to signals the GEMS ECU expects.

Having had some background in tinkering with 2000's VAG cars, i know that their Motronic 7.5 ECU's contain a 512x1 table which contains the MAF calibration in a very simple format: kg/sec against voltage. As such i have curves available for many common VAG MAFs. Those VAG MAF's are ubiquitous, readily available, and cheap new. I also already have a few units knocking around i can use for testing. Half the problem is thus "sorted"

If we can acquire the same curve for the Lucas 20AM, we can then create a translation table between the two.

Trying to find this information online seems pretty much impossible. Tuning info on the GEMS is rare, and i havent been able to find the factory calibration anywhere. However when replacing my broken MAF with yet another used one recently i noticed something potentially useful. RAVE provides airflow figures that should be seen at various engine speeds. One reason i know my newest-old MAF is junk is that it massively overreads against these figures. However while staring at the Nanocom, i realised it was showing both airflow in kg/sec and MAF voltage...

My first attempt then was to drive around, and use the Nanocom logging function to log these two values. Unfortunately Nanocom updates extremely slowly and seems to only pull one value at a time, which means the displayed airflow and voltage dont line up in time, and thus its all a bit random. Perhaps with many miles of logged data, we could get something resembling a curve, but i wasnt happy with this approach at all.

Spent some time thinking about it, and threw together a MAF "stimulator". Essentially its an Arduino Nano, and a MCP4725 DAC. The arduino is programmed to step the DAC output from 0v to 5v in 0.3125v intervals. Fishing about in my box of parts i found a matching connector for the harness MAF plug, so i can plug this contraption in, in place of the MAF. The idea being that instead of driving around, i can just sit on the drive, ignition on, engine off, and feed in the sequence of voltages and log the output.

So, tonight, i went to give it a try. And, provisional results are that it worked, kinda! Nanocom kept disconnecting, and i realised the battery was flat, which potentially was causing the nanocom weirdness, however i managed to capture a bit of data before i gave up with the battery down to 7 volts and went and fished out the battery charger instead. I also managed to overwrite one of my log files in the faffing. Nanocoms interface is pretty awful!

Interestingly you can see the lag i was talking about in the log due to how the nanocom polls each value. In order to try and minimise this, i'd set the Arduino to step the voltage once every 10 seconds. It appears to update the voltage first, then airflow shortly after, so on a few points you see the voltage steps up but the reported airflow doesnt change.

As you can see it was a struggle to keep the thing connected for more than about 30seconds, but you can also see some nice clear data points. Will revisit later in the week when it has a charged battery and see if we can capture the full sweep.

Any pointers or input gratefully received!
 

[elbekko]

Too bad you're not trying to do this for the later Bosch system, as that one has the curve in the manual. GEMS doesn't seem to in its manual

[Aragorn]

Hi, and thanks! The Bosch units are fairly readily available i believe. They're still manufactured by Bosch and sold directly in the aftermarket, so you can buy a genuine Bosch MAF for £100 or so. Additionally, Bosch tend to use the same internal element across quite a few different cars, so even if the P38 specific bosch unit became unavailable, you'd be able to find something that used the same core and swap it. (for reference the Thor P38 uses F00C2G2029 as its sensor element, used in many BMW's of the era)

As a result the situation for Thor models is much less of a problem. Unfortunately the GEMS/Lucas 20AM was used on early P38's and some Jags and thats it. The only aftermarket offerings are cheap knockoff units which have a very poor reputation.

A neat solution would be to use a Thor MAF for this swap, as that would enable an original RR MAF/airbox etc to be used without any modifications to the pipework/airbox etc but i dont have the curves for the Thor MAF. I've had a look at the graph in the page you've linked (super useful, thanks!), and oddly it doesnt look like a "normal" bosch MAF curve. Every Bosch MAF i've played with registers 1v at 0 airflow, and actually has the ability to measure negative flow in the 0-1v range. The graph in the manual you linked seems to show the curve starting from 0v however. Its also not detailed enough for what i need. Certainly however, this is something i can look into further.

If i could find ECU definition files and a rom dump from the Thor ECU, it might be possible to dig out the calibration table, but i dont know if the older ECU's use the same format. My concern ofcourse is the more "fuzz" that ends up in the data, the less accurate the whole thing becomes. I'm fully confident in the accuracy of the VAG data, and have a few VAG MAF's on hand. So i'll start there and see how i get on!

[Anderzander]

Hey welcome back ! 

[TD5toV8110]

Maybe talk to Mark Adams of Tornado Systems.

Very knowledgeable about all things Rover/Lucas.

He may have the curve you need.

Tornado Systems | Engine Management Technology Specialists

 

[Aragorn]

8 hours ago, Anderzander said:

Hey welcome back ! 

Thanks

[Aragorn]

8 hours ago, TD5toV8110 said:

Maybe talk to Mark Adams of Tornado Systems.

Very knowledgeable about all things Rover/Lucas.

He may have the curve you need.

Tornado Systems | Engine Management Technology Specialists

 

I have a recollection that i emailed some prominent GEMS tuners a few years back about this (this idea first crossed my mind when i bought the P38 some ~5 years ago) and got nothing. I dont recall who or wether they replied and couldnt help, or just didnt answer, i just remember it didnt move forward, and i put a used MAF on the truck instead. I'll have a check back thru my emails, but tbh i dont really want to point fingers. The whole tuning industry is a bit weird in general (and i'll stress i'm not aiming this specifically at Mark), no-one wants to share info and it all just ends up a bit of a horrible closed shop. You also end up with sharks selling dodgy incomplete or even downright incorrect data for money as a result.

I want to do my best to keep this open, hence posting the thread and i'll keep it updated with the findings.

If someone with "inside knowledge" wants to chip in and confirm my findings or help in some way then great. Otherwise i'll continue with my "plan" and see where it takes me!

[TSD]

I've not much experience with MAF sensors, but if you know the care and feeding of both known and target sensors, e.g. supply voltage, output type (voltage/ frequency/whatever), then maybe you could plumb the engine intake with both sensors in series, then just hoon around the neighbourhood datalogging both sensors at the same time? If they are analogue sensors, then I think you'd get enough of a curve to be able to fill the rest by eye. If both curves have similar shapes and ranges, then it could probably be made to work with either an analogue or digital 'converter' circuit.

The arduino could be programmed as an analogue to digital converter, dumping readings to a laptop via usb. It's not the best resolution ADC in the world, but it should be sufficient to show if the idea has legs, before spending any cash on producing a perfect solution.

I can think of plenty of reasons this wouldn't work, but not many reasons not to try it

[Aragorn]

That was also something i considered, the problem is to do that you need a known good perfectly working 20AM, and i'm not sure those exist any more  I also know for sure mine is knackered, as its registering about 6 times the airflow at idle than it should. Its also the fourth or fifth used one i've bought, and they all eventually go the same way.

Finding the Nanocom values was a bit of an eureka moment, as its showing me exactly what the ECU thinks a given voltage means in airflow. I just need to iterate the process i started above and capture all the values.

I've only got a few points so far due to the limited time i had yesterday, however its already producing a curve that looks at least somewhat sane (some of the values are a little off because i pulled them from some notes i made while testing the current MAF):

 

[TSD]

I don't think you really need to rerun that test, clearly the data you have can be represented pretty well by the polynomial shown. Any error looks to be small enough that neither driver nor ecu will notice.

Comparing that curve to the one elbekko linked, they seem to have very similar gain (slope of the curve) at high flow, but the GEMS sensor seems much more sensitive around 1V where I'd expect idle to be. Maybe that's why they are problematic, not because the sensor drift itself is bad, but the high gain means any drift in output voltage represents a larger drift in indicated air mass. It might be interesting to test a bunch of 'failed but working' sensors to see what the typical changes in the curves are.

I'm wondering if the change in the failed sensors is largely the offset voltage, so your curve above moves left/right, then that could be corrected by a very simple circuit. Unlikely, but I don't know what the real failure mechanism of the MAF sensor is.

[Aragorn]

I would like to get some points near full scale just to check, but yea, its already pretty close. Might not have time to touch it again until the weekend, will see how it goes.

The failure methods are varied. Two i've had have died when hot. The car would start and run fine from cold, after 10, 15, 20mins driving when the under bonnet temps get up, the engine will start coughing, spluttering and becomes completely undrivable. Unplugging the MAF fixes it immediately. I've had two where even from cold the car started but was chuffing black smoke out the exhaust like a diesel just trying to get it to idle, and was running extremely rough and again pretty much undrivable. The used one i've fitted most recently seems to read high but does at least allow the car to be driven. I presume the ECU manages to correct it so far with the fuel trims, but will eventually run out of "trim".

Im not really interested in trying to "correct" a broken 25 year old MAF. I want to be able to replace it with a new high quality modern unit. And hopefully some others can benefit from the efforts.

The engine will actually run without the MAF connected, and drives pretty well, infact you almost wouldnt know (though i'm undecided if thats a function of how bad the "working" MAF is, or just excellent default programming on landrovers part...), however its quite hard to start with it missing. It takes a moment for the ECU to realise the MAF isnt functional and to switch to whatever default program it has, and it needs a tiny fettle of the throttle to coax it into life. Too much throttle and it'll flood, not enough, and it'll fire then die out. Unfortunately while i've got the "knack" to get it going the wife doesnt, and the first time she drove it without the MAF she ended up stuck in the nursery car park unable to get it started, which as you can imagine went down like a lead balloon.

 

[elbekko]

On 5/3/2022 at 11:53 PM, Aragorn said:

Hi, and thanks! The Bosch units are fairly readily available i believe. They're still manufactured by Bosch and sold directly in the aftermarket, so you can buy a genuine Bosch MAF for £100 or so. Additionally, Bosch tend to use the same internal element across quite a few different cars, so even if the P38 specific bosch unit became unavailable, you'd be able to find something that used the same core and swap it. (for reference the Thor P38 uses F00C2G2029 as its sensor element, used in many BMW's of the era)

As a result the situation for Thor models is much less of a problem. Unfortunately the GEMS/Lucas 20AM was used on early P38's and some Jags and thats it. The only aftermarket offerings are cheap knockoff units which have a very poor reputation.

I realise that, mainly meant that it would've been a whole lot easier if the curve had just been in the manual like it is for the Bosch

 

[Aragorn]

Had another go last night, and it was a frustrating hour fighting with the Nanocom. Seriously unimpressed with that POS tbh.

It refused to stay connected to the ECU when displaying the "AIR - IDLE" tab. you'd get 20-30 seconds at most before it would just stop updating. To get it to reconnect, you had to exit all the way out to the main screen and go all the way back in again. and every time you did that, nanocom stops logging. In the end i just used my phone camera to record the screen, and manually plucked the values out later by rewatching the videos later while sat in front of the computer. Ofcourse each time you start logging its in a random place in the sequence, so you need to do it multiple times to actually see all the figures.

Managed 14 of the 16 data points the arduino was outputting which when plotted in excel makes a very nice looking curve.

I then used the "trendline" function to plot a line and used the resulting equation to produce 256 data points from the curve. its not 100%, but its very close and i'm sure its good enough for an old V8! So we're getting somewhere. I'll post some of the data in a little while.

 

[Aragorn]

Missing values are 2.5v and 5v

i just could not get the nanocom to capture the 2.5v point. I might reconfigure the DAC to output a constant 2.5v and see if that will give me a reading. I did get multiple readings for the 5v point, but it displayed 788kg/hr, which doesnt fit the curve at all. I suspect they've set full scale around 4.8v and anything beyond that is probably used to indicate a fault. In any case, 777kg/hr is around 270hp worth of airflow, so i dont think it'll ever be going beyond there in normal operation.