True, hopefully, but very doubtful Tom

Simon.

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240

Not a very succesful day today.

Did the filters and the oil and oil filter and the oil pressure sensor and tried to connect to the ECU. No communication. Tried various settings and couldnt get it to work.

Went for a drive and the issue is still there - kickdown at around 3000-3500 revs and if you get past this it seems to smooth out.

Was offered a chance to get the tyres stripped from my wheels for free (Having them blasted and coated this week) so put the jeep on axle stands to do this.

After we had finished and had some tea thought to check the software I had downloaded - I had downloaded XXI-N. Downloaded the correct software and surprise surprise it connected. Took some screen shots of the jeep whilst it was runnning:















The Tacho on the jeep and the RPM on the Piro are pretty much identical. Also saved the configuration in a configuration file and can email that if that is of any use.

Tom

You may have a slightly dodgy map sensor, nozzles in injectors could probably do with being .5mm larger with a bit less gas pressure, but I doubt any of that's the problem.

Clear to see a possible issue in calibration - The numbers in boxes represent how much the ECU multiplies petrol injector duration to arrive at gas injector duration, gas ECU also adjusts gas injector duration to compensate for gas pressure and temperatures (you had fairly cold gas at time of screen shots else gas injector duration would have been a bit higher). I said (previous post) those numbers should fall towards the bottom of screen (where engine load is greatest), we can see at low RPM this is the case in your calibration... But looks like someone has made it far too rich in all boxes from 2000rpm with 10ms load to 6000rpm 18ms load. Would bet someone has been in and progressively added in effect 28% more fuelling in those boxes, which might have already been set too rich...

You've already saved this calibration, so could reload it to get back to the same state as you're in now.

Only a guestimation but try this: On Bank1 table copy from 1500rpm 10ms to 1500rpm 18ms, then paste into 10ms to 18ms boxes all the way across from 2000rpm to 6000rpm, then copy the entire bank1 table and paste over the bank2 table... Calibration still won't be nearly correct (can see other issues in map) but I'd expect the engine to run much better at that. If you'd rather, you could email me the file, I'd edit it and email back.

As said, wouldn't be too surprised if there's an issue with an injector or two of this age too, this isn't and likely wouldn't be identifiable from screen shots but could cause some cylinders to run richer/leaner than others at certain rpm/load combinations.

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240

Thanks Simon. Will give the values a try today. Hopefully will have the wheels on in a week or so so can try the car out again.

Where do I get different size injector nozzles from?

And the MAP sensor - that's the original Jeep one or is there one for the LPG too?

Thanks

Tom

TinleyTech, other suppliers and some installers will stock OMVL injector nozzles and AEB025 pressure sensors, best to describe your injectors or presumption might be you have the later OMVL Superlights. Changing nozzles or adjusting reducer pressure would mean you'd need to do a full calibration though, the numbers in your calibration would all need changing, so probably best left alone for now. I'd rather see a lower pressure on Prio because system doesn't seem to correct for pressure as well as other systems while looking at your system pressure gauge it is almost at the end of it's scale (though pressure will be lower under engine load as Dream reducer isn't very pressure stable).

Was referring to the LPG system pressure sensor, it incorporates both a map sensor and gas pressure sensor. There may be an issue with it because it is unlikely your manifold pressure is 0.23 bar at idle, would expect more than 0.3 bar... But sensor might be OK, in which case it may be reading incorrectly due to an ECU related issue such as bad earth - Also noted your system voltage reads only 12v with engine running and temperature readings are low.. I should have asked how warm the engine was and how long it had been running on gas when screenshots were taken? May not have any consequences anyway and could just be another of Piro's quirks.

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240

Sorry should have said - the LPG had not long switched over - I was pretending to drive the jeep whilst it was sat on axle stands....

I haven't done anything today - been busy sorting out some of the mechanics in the jeep. Seemed sensible to concentrate on that whilst I have no wheels!

I think I should have the wheels back in a week or so so will be able to drive the jeep, get it good and warm and get some decent readings.

I think I'll pop up to you - I have some time off over xmas and new year so then would be good depending on your plans.

Thanks for your help Simon, it's great having someone willing to share their knowledge.

Tom

You're welcome, Tom.

No probs you popping up over Xmas.

Just one thing - screen shots show system was running on gas, amber light on switch is on and gas injector readings show the pulse duration...

Edit -Don't know how I missed it before, I read 'had not switched over', you said 'had not long since switched over'.

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240

Was hoping to have my wheels back this weekend so I could have a drive and see how the changes you suggested affected things. Unfortunately I was a day late getting the wheels to the Enamelers and as a result not ready.

In the meantime, even in neutral you may notice a difference revving the engine.. If you got it to 3000rpm and momentarily blipped the throttle before it may have hesitated, doing the same now it may not.

As implied/said above, even if the misfiring/hesitation now seems cured would still advise having it set up properly.. It is unlikely your calibration is anything like correct.

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240

This sounds like the firmware problem that the Piro system suffered from.

If this issue has been there since you have had teh car then you might be able to solve it by updating the firmware to the latest version. Updating the firmware solves the problem.

Under certain load conditions the ECU does not drive the injectors correctly in the 3-3500rpm range

_________________
http://www.classicswede.com/
http://www.classicswede.co.uk/LPG/cat17 ... 10800.aspx

LPG installer Anglesey North Wales

DIY LPG Kits

07824887160

Interesting.

At idle and with no load there has never been any problem with fuelling - so sat on axle stands I feel no difference at all.

I think flashing firmware at the moment is a step to far - I have a system that works (to a fashion) and id rather not have a brick.

Out of interest what is the latest firmware version as I'm guessing I should be able to se that with the software?

Tom

Hmm interesting. Tried it today on the axle stands and there is a hesitation there which was not there before. Only when revving hard from tickover - when increasing revs gently it's fine.

I'm pretty sure I didn't feel this with the original settings - still have the file so might put the old file back in.

You will not be able to get it into the problem area without load.

You can see the fault with a scope (not the software one) but it is not easy to see. With the highest level pass code you can make it better by adjusting the peak and hold settings but it does not fully cure it.

You look to be running the latest software so it will have the latest version of firmware on it

I did all the UK tech support for this system and this is a fairly common problem that could not be fully cured untill the latest software and firmware came out but by that point the problem had been there for so long and further development of teh system was abondond as the installlers had lost faith in it.

The pressure sensors also dropped off like flies, you can replace with a AEB025 but the pipe connections are the opposit way around

_________________
http://www.classicswede.com/
http://www.classicswede.co.uk/LPG/cat17 ... 10800.aspx

LPG installer Anglesey North Wales

DIY LPG Kits

07824887160

You wouldn't notice any difference while the engine is working at any point outside the areas that you adjusted in the calibration tables - the major changes made were to 3000rpm+ medium to high load settings.

You do seem to be seeing a difference (for the worse) when revving the engine in neutral (i.e. in the range you adjusted). However, as said before, those numbers should really get lower towards the bottom of the screen if the mapping is at all correct - Would expect the adjustments I suggested to improve matters but only if the ECU was working properly under such engine operating conditions... It is looking like the Piro issues Dai and myself described is likely the reason for the unintuitive numbers in the map which I suggested you change, so the real fix may be updating the firmware as Dai suggested whilst sticking with the changes I suggested...

If you don't fancy updating firmware yet, then before reverting to your previous settings you might try making changes to the 'acceleration' sliders.

Again, your best bet of sorting this and knowing all is correct may be to be visit someone who knows this system, it is possible to get seemingly good results in terms of drive-ability etc while mixture problems are still present, which could cause engine damage in the long term.

Side point, you earlier described your interior fuel switch as same as one you'd seen online... It looks almost identical, has same connector, works the same, but most switches of that design are not compatible with Piro (or at least Vice/versa).

Edit 'cos Dai's and mine posts almost crossed (Ayup Dai!), as Dai says you won't be able to get it to stay in the problem area, you may be able to get it in the problem area for only a split second until rev limiter kicks in, but not if rev limiter prevents you revving that hard in neutral (plenty of vehicles won't let you rev past maybe 4K in neutral).

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240

Well had a play today.

Seeing as i have saved copies of the configurations I updated the firmware in the ECU. All went well with no bricks.

As i was on a roll I let the system carry out an autotune. It went through the various process's - one interesting point was that it was difficult to hold a steady throttle opening - even when on petrol. That needs some investigating. Maybe the throttle cable.

Autotune completed and I ended up with the following screen afterwards. It seems the system is suggesting the injectors are too small? (Simon, you suggested this to start with?)



Tom

Autotunes normally work by changing all the numbers in the table by a percentage, Piro does the same but may change all the idle rpm cells by a percentage separately to the off idle cells. I dunno if Piro autotune resets the cells in the map when it is first run, if not then the same errors in the table would still be present. If autotunes did a good job at sorting mapping issues for us we wouldn't need to learn mapping skills, but they don't.

I did suggest your nozzles may be too small, that may still be the case but isn't compounded much by the autocal warning - the same warning would appear if this system were fitted on some engines even if nozzle size was correct, it is generated when multiplier (number) in cells used during autocal are high but on some engines (those with low petrol injector duration at idle such as engines with turbo spec injectors, which isn't always the same as engines with turbos, or high revving engines, or some with variable cam aspects) we'd expect a high multiplier in those cells even if nozzle size and pressure were correct. Better to manually check nozzle size and pressure are correct than take much notice of autocal error reports, my comment on nozzle size based on experience and the fact your engine idles at 3ms+ petrol injector duration. The point we (and autocal) hope to achieve is pressure and nozzle size set so that at full engine load gas injector pulse time is very similar to (would be) petrol injector pulse time with correct mixture. Autocal, amateurs and even lots of pros get that wrong, in fact autocal doesn't even attempt to adjust medium high load areas (and even on systems with more 'advanced' autocals they still get calibration wrong). Dai and I might not even use autotune, it is quicker and more accurate to effect the same changes by manually adjusting the numbers in the table, and we have a good idea what the shape of the table (map) should be while autocal doesn't. We often approach tuning the other way around by roughly setting low and mid range settings (so engine will at least run OK under those conditions) and then immediately concentrating on the high load areas to make sure we got the basic hardware setup (nozzles and pressure) right before properly tuning the mid and idle settings. When calibration is done in usual fashion (starting at low load points before moving onto high load) there's no easy way of knowing (though through experience we can anticipate) correct pressure and nozzle size for full load, and where that is wrong you could end up needing to repeat the usual fashion process several times.. This is where a lot of installers fall short, because having got a nice idle, decent fuel trims and correct fuelling at full load (but not at high rpm) they don't then take much notice of gas injector duration versus petrol injector duration at heavy loads at high rpm or do take notice but don't bother adjusting pressure and/or nozzle size and then starting calibration again. Where gas injector duration is less than petrol injector duration we might expect less accurate fuelling and problems with hesitation on acceleration / where gas injector duration is much higher than petrol injector duration we might expect the engine to run lean due to physical inability of system to provide enough fuel and maybe hesitation in acceleration in extreme cases.

Of course, no physical or software setting adjustments would get the system running 100% if the ECU doesn't do what it is supposed to but it may be possible to work around certain types of ECU issue such as problematic pressure correction. You've updated firmware to latest (comes with your software), so have done as much as possible to overcome ECU issues, so the only other easy things you can change are calibration, nozzles and pressure, injectors if necessary. There will still be Jeeps I converted using Piro running great on LPG.

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240

Thanks Simon, makes a lot of sense. I figured I may as well have a play as I had nothing else to do.

I often have to tune PID loops and for a lot of similar reasons I manually tune even if the controller has a good auto tune function as there are some thugs it just cannot take into account.

I'm still planning on coming to see you - but given the distance if I can get rid of a lot of the simple stuff (throttle cables for example) then hopefully we will have a productive day and not have to have a second trip.

Once I get the wheels back and know I have a working jeep I'll be in a better position to plan this.

Tom

Of course it's worth having a go and your experience may be handy.

Hope you get sorted, keep us informed and let me know if/when you'd like to come.

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240

From memory the autotune does rest the map when you start.

I dont use autotune and have not for a long time as it is faster to manualy adjust to where it needs to be.

You probably will need to go bigger on your injector nozzles

_________________
http://www.classicswede.com/
http://www.classicswede.co.uk/LPG/cat17 ... 10800.aspx

LPG installer Anglesey North Wales

DIY LPG Kits

07824887160

If he were closer to you Dai, I'd be advising he came to you mate.. Topcat, Dai knows his tech stuff very well, I was lucky enough to have Dai as supplier / mentor back in the day. The 'Piro will take over where 2568 left off' wasn't quite realised though eh mate!

I've got a Piro 8 cyl ECU in the garage which I installed on a Jeep 5.7 probably nearly 10 years ago, owner said he left vehicle stood for a year and when he put it back on the road the LPG didn't work, switch didn't even light, so I ended up removing Piro and fitted OMVL or some other AEB2568, but I've just thought... Does Piro have a relay for ign on that might have had it... Piro ECU I have might not be too difficult to fix...?

I can't even remember if Piro software and cable driver run on Win7? This new laptop came with Win8 but I immediately removed that and installed a multi boot system with Win7 64 bit, Win7 32 bit, XP 32 bit and DOS.. Only had this laptop a month or so but with various LPG software have needed them all since getting the laptop!

Dai, what do you reckon to his map and my suggestions? That 1.62 1.54 1.55 1.52 shape seems unintuitive for the bottom end, would expect a possible peak at 2ms, a low point at 2.5ms, rising to peak at 4.5ms, quick tale off and then gradual fall with load, the figures from 2500-6000rpm / 10ms to 18ms particularly wrong unless in attempt to compensate for Piro's failings, idle point probably not far out at 3.4ms on petrol, so mid load points probably also not far out. Dai and I have been discussing this sort of techy stuff in phone chats for many years, this might be a bit like a phone chat, only we're on a very loud speaker phone. I'm not calling now Dai else we might be talking til 3am

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240