M20B27 on LPG - Developments

[CHR1S1990]

LOL. I was only thinking that before on the way home from uni. With the BTB, 325i TB, M30 AFM etc. What I would say is to start with a 2.5 regardless of the difference in insurance - when you start modding the m20 in the way I have it makes it more track than road worthy, all the power is 4k onwards. Makes a lovely sound though

[daimlerman]

Rev limit is only approx 4500 so would hopefully show up any restrictions less.

Seem to remember that this is the peak power on the stock eta engine...
Just swapping the restricted 200 head for a 731(ex 320/520)will allow the engine to rev to the usual M20 limit,with no serious loss of low-down torque.

Upper rev limit will then be controlled more by your choice of inlet and injectors.

[StuBeeDoo]

If you are using an AFM opener is there any easy way you could try it without to see if there are any noticeable running problems?

If the BLOS is mounted before the AFM, you won't need a flap opener. If it's after the AFM, you will need the flap opener - which will leave you with the same problem I had with "passive" ignition timing.
I'd think that Chris's pictured set-up is pretty-much the only way to get BLOS to work with an AFM.

Chris - Why don't you just pull the fuel pump fuse??

[CHR1S1990]

I suppose i like the idea of a "safety" button at hand stu. A while back i had to boot it out of a junction near my house 'cause i was late for uni and traffic was terrible, and i had only been on LPG for about 3 minutes and my vap froze up. The petrol button saved me looking like a big fool and also saved me time! Id also like it there in case something went wrong on the motorway for example. In the summer i would have no problem pulling the fuse out. In fact, i think i could safely say the same about winter now that i have the BLOS fitted and now that I have fitted larger pipes to the vap.

[StuBeeDoo]

Stu, your findings with mixer size with AFM and MAF are exactly as I'd expect.

Do you reckon I should fit the 34mm venturi until I can afford to throw money at a BLOS then Alex?

[mr_dink]

Rev limit is only approx 4500 so would hopefully show up any restrictions less.

Seem to remember that this is the peak power on the stock eta engine...
Just swapping the restricted 200 head for a 731(ex 320/520)will allow the engine to rev to the usual M20 limit,with no serious loss of low-down torque.

Upper rev limit will then be controlled more by your choice of inlet and injectors.

Thanks, I'm aware of that and even have a b20 head in stock. Not planning any upgrades on the e28 though as I have an e30 in bits that I'm (very slowly!) converting to m50b25. That will be my speed/fun car and the e28 will be tuned for economy, Which is where the LPG comes in.

If you are using an AFM opener is there any easy way you could try it without to see if there are any noticeable running problems?

If the BLOS is mounted before the AFM, you won't need a flap opener. If it's after the AFM, you will need the flap opener - which will leave you with the same problem I had with "passive" ignition timing.
I'd think that Chris's pictured set-up is pretty-much the only way to get BLOS to work with an AFM.

Thanks, on the e28 it looks like there's plenty of room around the airbox area, so I will see if I can use a bigger airbox and get the BLOS inside somehow.

[e301988325i]

Stu, your findings with mixer size with AFM and MAF are exactly as I'd expect.

Do you reckon I should fit the 34mm venturi until I can afford to throw money at a BLOS then Alex?

With my 36mm mixer the stepper is doing quite a lot of work. You'll get better economy running a smaller mixer IMHO.

Chris's feednback about the BLOS is really promising, I can't wait to get mine on. I believe it starts well on gas with the BLOS because the BLOS trap door thing creates a vacuum behind it.

[StuBeeDoo]

Time for a mini update.

I fitted the 34mm venturi last weekend. Yesterday I did a 130 mile run and the economy was better, but I don't like to go off just one run.

Today I've fitted the LC-1. I've been up the road and the stepper is a lot more stable than I've ever seen it. Looks to me like it's wideband FTW if you're using Leo, or some similar closed-loop system.

I've ordered a BLOS, so that will be the next LPG-related job. On the petrol front I haven't done anything more yet, but now the LC-1 is up-and-running I need to get it running properly as at the moment if I should run out of LPG I can't even limp to the nearest station.

[StuBeeDoo]

I meant to say in the above post, but forgot, that IMHO anyone buying a W.A.R chip will either need;-
a) an LC-1, or similar
b) unlimited cheap access to a dyno
or
c) deep, well-filled, pockets to pay for loads of dyno time at the going rate.

[e301988325i]

More positive feedback on the BLOs, although the guy did have a terrible setup in the first place (25mm mixer 3.0L V6).

http://www.lpgforum.co.uk/viewtopic.php ... c&start=45

[StuBeeDoo]

I've ordered a BLOS, so that will be the next LPG-related job.

The BLOS arrived yesterday. I'm hoping to get it up-and-running later today.
Interestingly, in the instructions it says that a stepper motor can be used. Not sure I'll have time to do much experimenting with the Leo this weekend, though, so to start with I'm just going to lock the default at 255 and let the BLOS do all the work.

[StuBeeDoo]

OK, so the BLOS is fitted and the car drives on it. Still need to do some more tweeking to the fuelling, though.
I just hope that it will start from cold as I need the car first thing in the morning and I haven't yet had time to get the bloody thing to run on petrol.

[e301988325i]

How are you getting on Stu?

[StuBeeDoo]

How are you getting on Stu?

Started better than I expected from cold this morning, but not great. I'm still having trouble getting the idling speed and mixture right. I've checked the ICV and it seems to work OK - opens wide when cranking, closes completely as soon as the TPS goes on to p/t and is just open slightly on idle.
Looks like I'm going to be chasing air leaks now.
On the Leo (I haven't got my head 'round the Innovate software yet) I've got a O2 sensor reading of 4.6-4.9v (ie weak) all the time except for under very hard acceleration, when it drops to around 3v (still a little below stoich, then).

This morning I was ready to throw-in the towel again, but I'm slightly more positive now.
What I really need to be doing is getting used to using the Innovate.

I'll hopefully have another hour or so on it tomorrow morning, before an enforced lunch and afternoon with the in-laws.



That's if I can keep myself away from my new hobby.

[StuBeeDoo]

On the Leo (I haven't got my head 'round the Innovate software yet) I've got a O2 sensor reading of 4.6-4.9v (ie weak) all the time except for under very hard acceleration, when it drops to around 3v.

OK, today's lesson is "Don't trust Leo read-outs"!

I've had the Innovate working this morning, and the AFRs aren't quite as bad as the Leo had me believe.

Currently I have an AFR of around 20 at idle, 17-18 pottering around town at 20-30mph (accompanied by a slight jerkiness), 15-16 cruising at between 50 and 70, and under hard acceleration with WOT it drops to 14.5-15.
Comments please......

I can't find any air leaks, but that doesn't mean there aren't any. I'm tempted to fit one of my spare inlet manifolds but I'm not convinced that the idle speed problem* isn't ICV related. I don't have time to do any more this week, so next weekend I'll be trying to get the petrol fuelling right. If I can't get the idle right on petrol, that should be a good indicator. I should have a spare ICV somewhere, but it's a T-shape so before I can try it I need to source some pipe the correct size.

*Idle speed problem......
I can't get it below 1000rpm at all. To do this, I have to weaken the mixture right off - as in both adjusters on the BLOS screwed right in. The idle is then starting to get lumpy. Set like this the car hesitates on the road and is holding back until 3k rpm.
Screwing either of the adjusters out so that the car runs smoothly on the road lifts the idle speed to anywhere up to 1700rpm.

At the end of the day, it needs to be running right on petrol first, doesn't it?


BTW, what should happen if the engine is idling and the ICV plug is disconnected??

[e301988325i]

Hi Stu, on my BLOS, I had to break through a small piece of the casting, about 0.75mm x 1.5mm throttle body side of the door, just behind the tapered slot piece. Is there any obstructions to the door closinf fully? does it move freely?

Have you adjusted the vap at all?

You've always sruggled with high idle, to check air leaks, with the engine idling, take off the oil fill cap and see if this makes a diffence to how the engine runs.

[Gunni]

On the Leo (I haven't got my head 'round the Innovate software yet) I've got a O2 sensor reading of 4.6-4.9v (ie weak) all the time except for under very hard acceleration, when it drops to around 3v.

OK, today's lesson is "Don't trust Leo read-outs"!

I've had the Innovate working this morning, and the AFRs aren't quite as bad as the Leo had me believe.

Currently I have an AFR of around 20 at idle, 17-18 pottering around town at 20-30mph (accompanied by a slight jerkiness), 15-16 cruising at between 50 and 70, and under hard acceleration with WOT it drops to 14.5-15.
Comments please......

I can't find any air leaks, but that doesn't mean there aren't any. I'm tempted to fit one of my spare inlet manifolds but I'm not convinced that the idle speed problem* isn't ICV related. I don't have time to do any more this week, so next weekend I'll be trying to get the petrol fuelling right. If I can't get the idle right on petrol, that should be a good indicator. I should have a spare ICV somewhere, but it's a T-shape so before I can try it I need to source some pipe the correct size.

*Idle speed problem......
I can't get it below 1000rpm at all. To do this, I have to weaken the mixture right off - as in both adjusters on the BLOS screwed right in. The idle is then starting to get lumpy. Set like this the car hesitates on the road and is holding back until 3k rpm.
Screwing either of the adjusters out so that the car runs smoothly on the road lifts the idle speed to anywhere up to 1700rpm.

At the end of the day, it needs to be running right on petrol first, doesn't it?


BTW, what should happen if the engine is idling and the ICV plug is disconnected??

Can you have your software show lambda?

If so keep it to about 1-1.03 for idle
1.0 for cruise and 0.9 for WOT running

What are you using for tuning petrol?

Does your throttle body have a bypass screw?
Is the TPS working right?

[StuBeeDoo]

Thanks for ^^^ that Gunni.
My Innovate will show lambda - and it can be programmed for lambda 1.0 at 15.5:1 AFR to suit LPG.

I'll also be using the Innovate to tune the petrol.

As far as I can tell, the TPS is fine. As for the bypass on the TB, I'm not sure - I'll have to have a look.

[StuBeeDoo]

Hi Stu, on my BLOS, I had to break through a small piece of the casting, about 0.75mm x 1.5mm throttle body side of the door, just behind the tapered slot piece.

Can you expand on this please Alex? As in why? Do you mean one of the vacuum holes?

Is there any obstructions to the door closinf fully?

None I can find. The angled screw is wound right off.

does it move freely?

Seems to.

Have you adjusted the vap at all?

As per the instructions for the BLOS I wound the 8mm adjuster right in, and I double checked that the 10mm one is between 14-16mm as per the OMVL spec.

You've always sruggled with high idle, to check air leaks, with the engine idling, take off the oil fill cap and see if this makes a diffence to how the engine runs.

I'll do this tomorrow and report back.

what should happen if the engine is idling and the ICV plug is disconnected??

Anyone???

[Gunni]

Thanks for ^^^ that Gunni.
My Innovate will show lambda - and it can be programmed for lambda 1.0 at 15.5:1 AFR to suit LPG.

I'll also be using the Innovate to tune the petrol.

As far as I can tell, the TPS is fine. As for the bypass on the TB, I'm not sure - I'll have to have a look.

Your innovate only measures mixture it can´t tune fuel as you know. So how are you going to tune fuel?

[StuBeeDoo]

Your innovate only measures mixture it can´t tune fuel as you know. So how are you going to tune fuel?

Miller W.A.R chip.

[e301988325i]

Hi Stu, on my BLOS, I had to break through a small piece of the casting, about 0.75mm x 1.5mm throttle body side of the door, just behind the tapered slot piece.

Can you expand on this please Alex? As in why? Do you mean one of the vacuum holes?

There is a tiny bit of casting wall that wasn't broken through on mine, the wall wa approx 0.1mm thick, and this is where the idle gas flows through into the air stream, if it's not broken trough then IIRC the idle adjustment screw has no effect.

Just in front of the narrowest piece of the slot in this photo, through the raised section,not the taper. It pushed through will a ball pen it was so thin.

[StuBeeDoo]

Thanks for that Alex.

I've done the oil filler cap test and taking the cap off, the engine runs like a bag of shit. No air leaks, then?

I tried a spare T-shaped ICV bodged-on and it idles better. Looks like I've been sold yet more "good working condition" parts, in the shape of 2x L-shaped ICVs, that aren't as described.
For all the frakking about I've had with this £250 for a new ICV looks a good deal. I'm getting to the point where I'll not bother with second-hand stuff anymore.

This afternoon I'll get the laptop connected again and see if I can get some sensible lambda readings and idle speed with the T-shaped and if it's OK I'll have to order some silicon pipe to get it fitted properly. Changing the intake boot for the correct one for the T-shaped ICV isn't an option because it won't fit with my BLOS.

[StuBeeDoo]

OK, I have made progress.

Idle is now a lot smoother and is at 1k +/- 100. I still need to do a little work on it but I haven't had much time today.
It's now running richer throughout the rev range and drives far better than it has for ages. I've got it to lambda 1.15 idling, 1 - 1.05 pottering 'round town and 0.9 - 1.0 under hard acceleration.

I'll have another go at the weekend, and I'll also be having a go at getting it to run on petrol as well.

I've found a source of new L-shaped ICVs for a sensible price, but for the mean time I'll be getting a silicon hose to connect the T-shape properly, rather than the 3 bits of hose clamped together I have at the moment.

The L-shaped ICVs are now in the bin. Another lesson learned. 3 different units, all caused the same symptom.

[StuBeeDoo]

I've now got the ICV connected with a 135deg bend silicon hose, but it's still not ideal - I should have got a 180deg one.
There is still a very slight problem with the idle speed, but I know that it's ICV related. It's no worse than it's ever been. It's always been there, and Brian commented one time that it's a known issue, so I'll live with it for now.
I haven't done any more with the BLOS as the engine is running reasonably well on LPG at the moment.
Next weekend we're due to do a 200-ish mile round trip so I'll hopefully get a better idea of what the LPG consumption is like. My first impression is that my commuting figure isn't really that different, but having said that I've been enjoying the improvement in performance.

Because I'm running the MAF again, I haven't been able to drive on petrol at all for quite a few weeks now. I've started mapping the W.A.R chip on the petrol fuelling and so far I've got it to idle, and drive at low engine speeds (up to about 2.5k rpm), so at least now I'm not totally stranded if I run low on LPG miles from a refuelling point. I'll be continuing with the mapping on a W.I.P basis, but it's likely to be a long, boring, task. It took me something like 4 hours to get to the point I'm at now. The Innovate has been a great help and TBH if anyone is contemplating getting a W.A.R chip - or any tuneable engine management - I'd have to say that it's an absolute must.

[e301988325i]

More pictures, I've resized them a bit this time.

[StuBeeDoo]

Just got back from a 190 mile cruise around Yorkshire. As I've said more than once before, I don't like to get too excited by consumption figures taken from a single tankfull, but I have to say I'm impressed.
With the mixer and any of my assortment of venturii (is that the correct plural??) I wouldn't have expected any more than 20.5-21mpg at the very most. What I got was...............

22.6mpg.


I'm tempted to set it all up with the stepper active, just to see what effect it has on the performance and economy. I'm also tempted to try running with the over-run cut-off as well.

Monday morning I need to do an Inspection 1, so while the BLOS/MAF/air filter assembly is off the car for the element change I'll double-check that my BLOS doesn't have the same casting issue as Alex's.

[CHR1S1990]

Looking good stu. Hard to believe that an open loop carb style kit can achieve better results than a closed loop, but it does! I must say, I think a lot of it (of course plus the more efficient design of the carb) is down to the accuracy you can monitor/tune the mixture with the AFR gauge. Best piece of kit i have invested in in a while!

FYI, i was getting 22mpg on normal driving before BLOS, I am now getting 24.2mpg and it is rising daily (litterally) - note I fitted a BTB3 at the same time, and my pedal has been to the floor since - looking forward to see what I can get driving economically. this is m20b20 for those who dont recall my car specs, so would expect to see something similar on an m20b25.

IMO, i think adding the stepper may be a step backwards. the leonardo will be trying to adjust the AFR based on the narrowband simulation from the LC-1 - i know how 'off' these values actually are. I wont be using the stepper system again tbh.

chris

[e301988325i]

I'm sure some of the increased efficiency will be coming from the reduced vacuum load is placed on the engine because of the restrictive fixed mixer. Bloody good news, how did your casting compare to mine chris, was it broken through?

[CHR1S1990]

Yes the hole seemed to be broken through alex so i left it untouched, tried it and it seems fine if a little lumpy at idle but a nice "narrow" range of AFR so im quite pleased! also note the above MPG is with my mixer adjusted to more 'rich' than stoich for LPG to benefit power more than economy so all in all its quite impressive!

im very pleased with the setup as it is, the next stage is fitting my 2.7 (nearly done) which should work just as well. Then megasquirt so i can properly bin off the AFM alltogether (though i may as well do that now as i dont even have a working fuel pump in!). Then i will need to be looking at injection LPG as the BLOS wont cope with the rotrex i plan on bolting to the m20 any success stories of megasquirt and LPG, BMW/m20 even better would be much appreciated.

chris

[StuBeeDoo]

My BLOS doesn't have the idle slot issue yours did Alex.

[e301988325i]

Well with two of yours as standard, I'm glad I did clean up and break through that little bit of the casting!

[CHR1S1990]

you got it up and running yet alex? My MPG hit an all time high of 25.0 today, and thats with short town/shopping trips since I finished university

how's your economy now stu? did you end up re-fitting the stepper motor?

[e301988325i]

I haven't worked on the car for months now, got so much to do to it, it's ridiculous. I'm back at college and the coursework is eating up all my free time, well for the next 4 weeks until the end of term. However got the cash saved up to get the 2.8 built, and fix the rust, piggyback, lpg etc, good news on the economy, that means it's more efficient and thus must release power!!!!!!

[StuBeeDoo]

I haven't done much else to my car and for the forseeable future won't be. I've lost interest in it and all I can be bothered to do at the moment is keep it on the road. I just can't be arsed anymore.
It's running well on LPG, but still uses more than other people claim for their M20s. Currently, over 600 miles of mixed commuting and runs, it's averaging 19.4. I've got to the point where it will start and limp to the nearest LPG source on petrol, but I'm still having issues with the W.A.R chip.
No, I didn't bother making the stepper active. The general concensus seemed to be that the drivability would suffer again.