Advice please!

Post Reply
Message
Author
Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Advice please!

#1 Post by Pinger » Fri Feb 21, 2020 4:13 pm

Hello all.
I've recently acquired a 1999 Chevrolet with the Vortec 5.7l engine and a single point with stepper LPG set-up and I could really use some advice as there are a few things about it that are not as they should be. I haven't driven it on the road as yet - only shunting in the drive way to effect repairs following an inlet backfire which became a crankcase explosion which ruptured the timing cover. That's all sorted now and the cause - LPG inlet to induction system upstream from the PCV inlet to crankcase...

Two areas of concern. Firstly I know it isn't running as it should on LPG as there is a recent invoice from a garage who were tasked with investigating the Lambda sensors, presumably due to EML being illuminated. They identified the sensors as operating correctly when running on petrol but not with LPG. They then ran out of LPG and went no further. So, clearly there is a problem there which needs resolving.

Secondly, the throttle butterfly is around 4'' in diameter. The 'mixer' is a reducing orifice with a rectangular cross section LPG inlet terminating at at its ID and that ID is circa 3''. A quick comparison of the areas suggests the mixer ID can only flow just over half as much air as Chevrolet intended so power output will be half what should be there. I can't see a reason to tolerate that!

Looking at it from that perspective and in conjunction with what the installer told me (''the reduced ID is to create a venturi effect'') it seems to me that beyond half the rpm of maximum power (circa 2300 rpm) the reducer is no longer acting as a venturi but will be throttling (in the truest sense) the air flow and creating lower pressure downstream than Chevrolet intended and changing substantially the pressure differential across the mixer that the LPG inlet sees. This I suspect is the root of the problem the garage was tasked to fix.

The questions I could really do with answering before I get stuck in are:
1) How and where do I monitor what is going on. I'm aware there is a hand held unit available that reads the Lambda sensor outputs, rpm, and stepper activity. Where do I obtain one and where does it plug in? Is there a version of it as software only that I can download and install on a lap top I have set up with Delphi software that connects to the OBD port? Can the above parameters be read from the Delphi software - assuming my software covers Chevrolet models? (I suspect so but not the stepper activity).

2)Why do reducers with single point systems always supply at slightly below atmospheric pressure? I'm aware of the 'mixer conundrum' where the size of mixer that affords full power leads to unstable idle and the mixer that stabilises the idle reduces airflow and thus Pmax. My thinking here is that I want to raise that pressure, bin the reduced ID mixer orifice and get the LPG flow at idle by raising (at the reducer) the LPG pressure rather than rely on venturi effect to draw the LPG in and then allow the stepper to control (reading from the Lambda sensors) the mixture elsewhere in the load/rpm range. But for a nagging feeling I'm overlooking something.

So, there's what I'm up against and trying to achieve. Will some of you please put me straight, help me get this engine running properly, and save my sanity!
TIA.

LPGC
Installer
Posts: 4007
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Advice please!

#2 Post by LPGC » Fri Feb 21, 2020 5:16 pm

Check / read manifold pressure at wide open throttle high rpm, it doesn't matter if you do it running on petrol or on LPG the result will be the same.. I think you'll find manifold pressure reaches atmospheric pressure with the 3 inch restriction on your engine anyway, in which case the 3 inch mixer won't really be causing a restriction so won't be limiting torque/bhp. If you look at hod rod forums etc they talk about running 58mm tb's on normally aspirated engines that produce over 350bhp and see no difference in power by switching to a wider throttle body.

1. There's the manifold pressure test as mentioned above. If the engine has a map sensor you might see that info in OBD live data using a cheap generic OBD2 live data reader tool, same tool will also read lambda voltages etc. Since the LPG system has a stepper motor it may have a diagnostic/setup port that you can connect a laptop to using an interface cable, depends on what type of LPG ECU controls the stepper. Since it has a stepper the LPG system must read lambda voltage too, so LPG software would be another window on lambda voltage (and stepper position and probably TPS).

2. Mixer systems rely on 'signal' (slight vacuum) caused by the venturi to suck gas from the reducer... Easier to think of the reducer supplying gas at atmoshpheric pressure and any slight depression sucking the gas as the reducer tries to maintain atmospheric pressure in the gas outlet - so the more signal the more gas will come out. You can't just increase reducer pressure as that will just have the effect of offsetting the signal to flow curve with the effect the mixture will be much richer at low loads but not much richer at high loads. Reducer signal sensitivity is a bit different to reducer pressure.

Assuming this ran well when originally fitted, rather than change the design why not try to find if any component(s) have gone wrong? Steppers can fail/seize, reducers can go wrong, lambdas can go wrong but you seem to have ruled lambdas out, electronics (ECUs) can go wrong.

What make are the components?
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

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#3 Post by Pinger » Fri Feb 21, 2020 6:30 pm

Thanks LPGC.
I intend (when I get it on the road beginning of March) to run on petrol with and without the mixer to see if the power is affected or not so that will be answered then. Heartening to hear that it probably wont be an issue.

The components are: Reducer - OMVL R/90 Super 8; ECU - OMVL RegMatic Millennium; and 2 x Pitagaro P1 Brev emulators; and a a AEB OBDii Fix black box that I cannot identify its purpose.
As far as I can make out so far, it reads Lambda sensors, TPS, MAF and MAP. Since I have posted I have found that my Delphi software covers the Chevrolet (and maybe Wurth WOW also) so I' should at least be able to read Lambda data. Where will the port be for a separate scanner to read stepper activity? Do Tinley Tech sell the scanners? (I' had a browse of this site while waiting for membership approval and I've seen the video of the data being read (and power valve being used) on a Rover so have some idea of the system).

Fair point re a fault somewhere - which adds to the need for diagnostics. The Lambda sensors being OK was as far as that garage got but if they are OK on petrol, for now, I'll assume they are healthy.
I did have it running briefly on LPG (low load, low rpm just shunting on the drive) until the LPG ran out (my mistake) and the backfire and crankcase explosion occurred and it seemed to me to be running fine but I wasn't paying attention to the dash for the EML. Maybe on open road running is where it plays up. Before I obtain more in the way of diagnostics, are there physical checks for the stepper? Can I physically move it to check it is unseized and if so, do I have to watch for it losing synch with the ECU?

I'm still struggling with this:
''2. Mixer systems rely on 'signal' (slight vacuum) caused by the venturi to suck gas from the reducer... Easier to think of the reducer supplying gas at atmoshpheric pressure and any slight depression sucking the gas as the reducer tries to maintain atmospheric pressure in the gas outlet - so the more signal the more gas will come out. You can't just increase reducer pressure as that will just have the effect of offsetting the signal to flow curve with the effect the mixture will be much richer at low loads but not much richer at high loads. Reducer signal sensitivity is a bit different to reducer pressure.''

I understand carbs, Bernoulli, the whole pressure drop thing - but the gas can be delivered under pressure negating the need for the pressure drop (the distinction between carbs and fuel injection). I'm not seeing why I can't lose the mixer orifice and introduce the gas under its own pressure and the ECU under Lambda guidance instructs the stepper to control mixture. What am I missing? Isn't this more or less how LPG injection systems work (albeit at higher pressure than I envisage)? Am I crediting the ECU and stepper with more mixture control capability than it actually possesses?

Gilbertd
Advanced Member
Posts: 8519
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Advice please!

#4 Post by Gilbertd » Fri Feb 21, 2020 10:03 pm

If the video you found was this one https://www.youtube.com/watch?v=b6sGp2I9lts, it's mine. 4.0 Litre P38 Range Rover running an identical system to what you have. You don't need a hand held scanner, you just need the cable (whcih Tinley Tech can supply) and the software (which will come with the cable) and a laptop to run it on. The backfire will have been caused by it running ridiculously lean due to running out of gas. Had you been driving at the time you would have felt it going all sick and flat with the amber LED on the switch slowly flashing to indicate a lean mixture. At which point you switch over to petrol and carry on.

Note the AEB175 is sold as the AEB Leonardo, the OMVL Millennium, The Bigas Pegaso and the Zavoli Regal and possibly a few other names but they are all the same unit. It only connects to the TPS (so the controller knows if the throttle is open or closed) and a lambda sensor to give feedback on the mixture and control the stepper. The Piagoras are emulators that are connected in place of the petrol injectors so the petrol ECU doesn't know the injectors have been switched off and flag a fault. Strictly speaking the OBD Fix shouldn't be needed but it is there to stop the EML from coming on when running on gas.

Your idea of raising the gas pressure and doing away with the venturi wouldn't work. Or it might at a certain revs and load but will never run at anything other than at that load and revs. Have a read of this http://www.diy-lpg.co.uk/articles/files ... -loop.html article which explains how and why it works and how to set up a system without the benefit of the lambda feedback closed loop system. The only difference in what you have is that it uses the output lambda sensor to fine tune the system. If you think the mixer is going to strangle it, try running it down a drag strip with teh mixer fitted and without, there will be virtually no difference at all, nothing you would notice under normal driving anyway. In fact, if you can advance the ignition timing to take advantage of the much higher Octane rating of LPG over petrol, you should be able ot get more power on gas not less.

The lambda sensor not giving an output on LPG could be because the calibration of the controller has been locked for some reason so it is unable to adjust the default setting on the stepper to keep things correct. You need the software and cable to look at the settings and you should be able to diagnose and correct any problems with that.
96 Saab 900XS, AEB Leo, sold
93 Range Rover 4.2LSE, Lovato LovEco, sold
97 Range Rover 4.0SE, multipoint, sold
98 Ex-Police Range Rover 4.0, AEB Leo, daily motor
96 Range Rover 4.6HSE Ascot, AEB Leo, my spare


Proud member of the YCHJCYA2PDTHFH club.

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#5 Post by Pinger » Sat Feb 22, 2020 3:26 pm

Gilbertd wrote:
Fri Feb 21, 2020 10:03 pm
If the video you found was this one https://www.youtube.com/watch?v=b6sGp2I9lts, it's mine. 4.0 Litre P38 Range Rover running an identical system to what you have. You don't need a hand held scanner, you just need the cable (whcih Tinley Tech can supply) and the software (which will come with the cable) and a laptop to run it on. The backfire will have been caused by it running ridiculously lean due to running out of gas. Had you been driving at the time you would have felt it going all sick and flat with the amber LED on the switch slowly flashing to indicate a lean mixture. At which point you switch over to petrol and carry on.

Note the AEB175 is sold as the AEB Leonardo, the OMVL Millennium, The Bigas Pegaso and the Zavoli Regal and possibly a few other names but they are all the same unit. It only connects to the TPS (so the controller knows if the throttle is open or closed) and a lambda sensor to give feedback on the mixture and control the stepper. The Piagoras are emulators that are connected in place of the petrol injectors so the petrol ECU doesn't know the injectors have been switched off and flag a fault. Strictly speaking the OBD Fix shouldn't be needed but it is there to stop the EML from coming on when running on gas.
Thanks Gilbertd - useful info. Where do I find the Millenium (the name on my ECU) diagnostic plug? Engine bay? I can't as yet find it - the nearest is a light grey 4 chambered plug about 2'' long but all it has is a wire coming out from each end and no additional contacts (though space for them) within. One wire appears to go downward toward the CPS or a sensor on the block. Doesn't really fit the brief. I've found the lead on TT's website, just holding back until I find the connector on the car.
I take it the Millenium ECU cannot be read via the OBD port?
Gilbertd wrote:
Fri Feb 21, 2020 10:03 pm
Your idea of raising the gas pressure and doing away with the venturi wouldn't work. Or it might at a certain revs and load but will never run at anything other than at that load and revs. Have a read of this http://www.diy-lpg.co.uk/articles/files ... -loop.html article which explains how and why it works and how to set up a system without the benefit of the lambda feedback closed loop system. The only difference in what you have is that it uses the output lambda sensor to fine tune the system. If you think the mixer is going to strangle it, try running it down a drag strip with teh mixer fitted and without, there will be virtually no difference at all, nothing you would notice under normal driving anyway. In fact, if you can advance the ignition timing to take advantage of the much higher Octane rating of LPG over petrol, you should be able ot get more power on gas not less.
I've read the articles (open and closed loop) but still not quite gelling!
The TPS - is it used in the processing for the actuator - or is it only there for the fuel transition? The Millenium ECU reads rpm. Is that (as above) part of the processing or for fuel cut off (safety) if engine stops? If the former in each case - I can more easily accept the limitations of the actuator - If the latter, well its is only going to trouble me if my mixer plate does actually inhibit power - and two are telling me it wont so for now - back burner. If you can answer those questions though, it'd be appreciated.
Gilbertd wrote:
Fri Feb 21, 2020 10:03 pm
The lambda sensor not giving an output on LPG could be because the calibration of the controller has been locked for some reason so it is unable to adjust the default setting on the stepper to keep things correct. You need the software and cable to look at the settings and you should be able to diagnose and correct any problems with that.
I don't know if there was no output or what was the problem. I'm guessing it was the EML (you mentioned the OBD11 fix to circumvent that - maybe a fault with it) but don't know as it was with the previous owner (I've just bought it and haven't been on the road as yet) but something prompted a trip to a garage in search of answers. It was advertised as having had the LPG system checked and was running fine on it. Included invoice says otherwise.

Gilbertd
Advanced Member
Posts: 8519
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Advice please!

#6 Post by Gilbertd » Sat Feb 22, 2020 3:57 pm

Find the Millennium ECU and look at the cable bundles coming out of it. The larger of the two plugs has 3 or 4 bundles of wire coming out of it and one of these is about 8"-10" long and has a plug on the end of it not connected to anything. It will either be a 8 way Molex plug (but with only 4 wires) or, if it is a later unit, a 4 way Supaseal plug. That is the diagnostic plug where you connect the cable to your computer. The Millennium cannot be read from the OBD port as it is a completely separate fuel system that only ties in to a couple of the sensors used by the petrol system for convenience.

The TPS signal is used purely to tell the Millennium if the engine is idling or running with the throttle open. You can set limits to the steps above and below the actuator default to different amounts at idle and out of idle, so it needs to know when the engine is idling. You can also set fuel cut off on the over run so it uses a combination of the TPS and RPM to decide when to shut off the fuel. The RPM input is used to tell it the engine is turning so if it stalls it immediately shuts off the gas supply but it is also used to tell it when to changeover. As an engine will run on LPG from stone cold, I've got mine set to changeover to LPG when the revs are descending at 1,100 rpm. Start the car (which will happen on petrol), blip the throttle to bring the revs to over 1,100 rpm and it changes over as the revs drop.

There are no limitations in the actuator, all it does is a bit of fine tuning depending on what the output of the lambda sensor is telling it. With an open loop system without the feedback from the lambda sensor you set the power valve (a fixed valve used in place of the actuator) to give good running at idle and out of idle.

I suspect the LPG system was running fine until it ran out of gas. That is the only reason you got the backfire, a ridiculously lean mixture. Where in the country are you? If anywhere near to me I don't mind checking the settings and calibrating it if it needs it. I'm on my fourth car fitted with a Leonardo/Millennium/Pegaso and wouldn't fit anything else.
96 Saab 900XS, AEB Leo, sold
93 Range Rover 4.2LSE, Lovato LovEco, sold
97 Range Rover 4.0SE, multipoint, sold
98 Ex-Police Range Rover 4.0, AEB Leo, daily motor
96 Range Rover 4.6HSE Ascot, AEB Leo, my spare


Proud member of the YCHJCYA2PDTHFH club.

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#7 Post by Pinger » Sun Feb 23, 2020 11:06 am

Gilbertd wrote:
Sat Feb 22, 2020 3:57 pm
Find the Millennium ECU and look at the cable bundles coming out of it. The larger of the two plugs has 3 or 4 bundles of wire coming out of it and one of these is about 8"-10" long and has a plug on the end of it not connected to anything. It will either be a 8 way Molex plug (but with only 4 wires) or, if it is a later unit, a 4 way Supaseal plug. That is the diagnostic plug where you connect the cable to your computer. The Millennium cannot be read from the OBD port as it is a completely separate fuel system that only ties in to a couple of the sensors used by the petrol system for convenience.
That's the bit of the loom I've been following and no sign of a connector. There are two severed wires though ( a black and a yellow). More investigation required. The install was done in 2002.
Gilbertd wrote:
Sat Feb 22, 2020 3:57 pm
The TPS signal is used purely to tell the Millennium if the engine is idling or running with the throttle open. You can set limits to the steps above and below the actuator default to different amounts at idle and out of idle, so it needs to know when the engine is idling. You can also set fuel cut off on the over run so it uses a combination of the TPS and RPM to decide when to shut off the fuel. The RPM input is used to tell it the engine is turning so if it stalls it immediately shuts off the gas supply but it is also used to tell it when to changeover. As an engine will run on LPG from stone cold, I've got mine set to changeover to LPG when the revs are descending at 1,100 rpm. Start the car (which will happen on petrol), blip the throttle to bring the revs to over 1,100 rpm and it changes over as the revs drop.
OK, so the TPS and rpm play no part in the metering....
Gilbertd wrote:
Sat Feb 22, 2020 3:57 pm
There are no limitations in the actuator, all it does is a bit of fine tuning depending on what the output of the lambda sensor is telling it. With an open loop system without the feedback from the lambda sensor you set the power valve (a fixed valve used in place of the actuator) to give good running at idle and out of idle.
..... so the actuator fuel trims rather than meters. This I think is where I was confused prior.

Gilbertd wrote:
Sat Feb 22, 2020 3:57 pm
I suspect the LPG system was running fine until it ran out of gas. That is the only reason you got the backfire, a ridiculously lean mixture. Where in the country are you? If anywhere near to me I don't mind checking the settings and calibrating it if it needs it. I'm on my fourth car fitted with a Leonardo/Millennium/Pegaso and wouldn't fit anything else.
The invoice to investigate the LPG 'fault' - whatever it was, predates me taking ownership of the car by a couple of months so something isn't as it should be and will need resolving once I identify what is wrong - maybe just EML illumination.
The back fire I put my hand up and accept responsibility for. I was told the tank was low on gas and I decided I'd use the last of it (to extricate myself from the wet grass I'd got stuck on). It stalled, and stupidity on stupidity, I restarted it. A moment later there was an almighty bang, much smoke from under the bonnet, and I'd later find it had shed most of the sumps oil. The back fire had ignited a crankcase primed with air and LPG as the PCV was drawing its 'air' downstream from the LPG inlet. Whoever configured it like that needs a slap. It goes without saying that that will be remedied ASAP. I'm relieved though to read your description of how running out of LPG presents itself when running on the road - ie, identifiable and with enough warning to change fuels before any ultra lean back fire can occur. I'm new to this LPG malarkey - yep, you can tell!
Thanks again Gilbertd - your input here is much appreciated. And thanks for the offer to have a look - alas I'm in the NE of Scotland (distillery country). With the advice I am getting here though, I will get this thing properly sorted. I have to - it's going to be my daily driver!

LPGC
Installer
Posts: 4007
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Advice please!

#8 Post by LPGC » Sun Feb 23, 2020 5:55 pm

It needs an RPM feed to know the engine is running or will shut off the LPG solenoids. RPM reading also makes for subtle differences e.g. speed of stepper movement is different between idle/off idle/high rpm. If you don't connect TPS some AEB175's will assume high load at above around 4500rpm regardless of load, in which case instead of aiming for lambda of 0.5v (for a 0>1v sensor) they'll aim for bit higher voltage (say 0.7v) for a bit richer mixture.

The positive pressure mixer system comments could give the idea you're thinking along lines of re-inventing a system that works like some of the Impco stuff or Etagas but in both cases the components have major differences to the more usual type of mixer system you have ;-)
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

Gilbertd
Advanced Member
Posts: 8519
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Advice please!

#9 Post by Gilbertd » Mon Feb 24, 2020 12:32 am

Pinger wrote:
Sun Feb 23, 2020 11:06 am
That's the bit of the loom I've been following and no sign of a connector. There are two severed wires though ( a black and a yellow). More investigation required. The install was done in 2002.
So it will be the early variant then. I've just dug my spare loom and controller out to have a look. From the larger of the two plugs on the controller there are a number of separate sleeved wires.
They are:
One carrying the power feeds, a Red wire and a Red/Black wire
A single Black wire (Ground connection)
A single Brown wire which goes to an ignition coil or rev counter feed to pick up the rpm signal
What appears to be a black wire but is actually the Blue/Yellow wire in a black sheath that goes to the TPS
What appears to be a black wire but is actually the Blue wire in a black sheath that goes to the gas solenoids.
Thicker black sleeving carrying the purple and grey wires for the lambda sensor connection
4 wires that go to the actuator (stepper)
and the remaining one is about the same thickness sleeving as the one to the actuator which has the diagnostic socket on the end. The wires in that are (on this one anyway) two pink wires, a White/Red and a Black). The socket may (should) have a protective cover on it so will look like a rectangular plastic block, it may not be apparent as a Molex socket it could be mistaken for a relay.

The only Yellow wires are a pair of wires that are connected to a relay contact that is open when running on gas that can be used for any other purpose that might need something disconnecting when running on gas.
96 Saab 900XS, AEB Leo, sold
93 Range Rover 4.2LSE, Lovato LovEco, sold
97 Range Rover 4.0SE, multipoint, sold
98 Ex-Police Range Rover 4.0, AEB Leo, daily motor
96 Range Rover 4.6HSE Ascot, AEB Leo, my spare


Proud member of the YCHJCYA2PDTHFH club.

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#10 Post by Pinger » Mon Feb 24, 2020 1:27 pm

Thanks Gilbertd - I've found it!
Exactly as you said, a rectangular plug with connections arranged in a line and the wire colours as you state. I can (once I get the appropriate lead) now access the diagnostic and programmable elements. I'l check the other wires you mention in due course (the weather here isn't conducive to that right now - damned snow!). I suspect Millenium access to the Lambda sensors is via the Chevrolet's own wiring. Each lambda sensor has what appears to be OEM wiring and, a section of the wiring from the Millenium's larger plug appears to connect with the Chevrolet's PCM (Powertrain Control Module) which AFAIK controls ignition and SFI (sequential fuel injection) - which is obviously in communication with the Lambda sensors.
The two severed wires are both (i think) yellow - in line with your assertion.

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#11 Post by Pinger » Mon Feb 24, 2020 1:46 pm

LPGC wrote:
Sun Feb 23, 2020 5:55 pm
It needs an RPM feed to know the engine is running or will shut off the LPG solenoids. RPM reading also makes for subtle differences e.g. speed of stepper movement is different between idle/off idle/high rpm. If you don't connect TPS some AEB175's will assume high load at above around 4500rpm regardless of load, in which case instead of aiming for lambda of 0.5v (for a 0>1v sensor) they'll aim for bit higher voltage (say 0.7v) for a bit richer mixture.
It's been a bit of a rollercoaster ride. To begin with, I thought I had a very simple open loop system. After talking to the installer and finding the actuator, I realised it is a closed loop system. From there I imagined the system read all the sensors the petrol FI system has access to and thus 'metered' from that data. I now realise that metering is a function of reducer pressure/bias, powervalve/actuator, and venturi action of the mixer. Unless that is also wrong. Other sensor data is merely for other levels of functionality - other than Lambda sensor for fuel trim.
LPGC wrote:
Sun Feb 23, 2020 5:55 pm
The positive pressure mixer system comments could give the idea you're thinking along lines of re-inventing a system that works like some of the Impco stuff or Etagas but in both cases the components have major differences to the more usual type of mixer system you have ;-)
You are correct. I realise now that ' positive pressure ' is inappropriate to a mixer system so will continue with the accepted practices. However there are a few other things I will be asking about in due course (eg, does raising the compression ratio to exploits LPG's higher octane # better suit the unalterable ignition timing set for petrol operation which is deemed to retarded as standard for best LPG running) and one thing still on my mind is why injection systems' injectors aren't just driven by the petrol injectors signals (with change over relays for each injector pairing) with flow matching via reducer pressure manipulation (as it is with petrol injection when fuel rail pressure is some value above MAP). The electrical impedance of each injector type is matchable (circa 1.8 Ohms) and having the full suite of data would surely improve metering. Unless much of that is related to temperature variation which LPG is less susceptible to. I understand that the injection systems crib from the petrol injector's signals but still feed them to an emulator rather than directly switching them to the LPG injectors.
I'll research the Impco and Etagas you mention - perhaps there are answers there.
Thanks for your input - much appreciated.

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#12 Post by Pinger » Mon Feb 24, 2020 2:03 pm

Gilbertd wrote:
Sat Feb 22, 2020 3:57 pm
As an engine will run on LPG from stone cold, I've got mine set to changeover to LPG when the revs are descending at 1,100 rpm. Start the car (which will happen on petrol), blip the throttle to bring the revs to over 1,100 rpm and it changes over as the revs drop.
Picking up on the above, why don't/can't you just start on LPG as well? Is it that the LPG won't deliver at cranking speed? If so, is there not a way around that (eg, tap into the vapour in the tank for starting only) or enablement from the reducer during cranking? If I could ditch my 35 Imp gallon petrol tank I could move the LPG tank from interior to where the petrol tank currently is without the aggro of having to install a reduced size petrol tank if I could get it to fire up on LPG.
Gilbertd wrote:
Sat Feb 22, 2020 3:57 pm
There are no limitations in the actuator, all it does is a bit of fine tuning depending on what the output of the lambda sensor is telling it. With an open loop system without the feedback from the lambda sensor you set the power valve (a fixed valve used in place of the actuator) to give good running at idle and out of idle.
I came across a video (I'll have to have another look for it) that showed setting up with a power valve and then 'matching' the actuator to that setting. Is that the procedure? How does the actuator know the power valve setting? There seems to be a preference for the actuator to operate over a tight an operating range as possible (though it can move further either way to trim). Is this merely a byproduct of a well set-up system?

LPGC
Installer
Posts: 4007
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Advice please!

#13 Post by LPGC » Mon Feb 24, 2020 3:45 pm

To answer your questions thoroughly would involve a lot of typing! If you search this forum you'll find they've already been answered many times, the answers are obvious anyway when you have a decent grasp of the basics.

In brief...

Injectors - Can't run petrol injection pressure to LPG injectors, petrol pressure is too high for LPG vapour injectors to open against, also with this kind of pressure in cold conditions gas vapour would condense back to liquid (which would really mess up LPG injector metering). Even if those points could be overcome you couldn't run gas injectors from petrol injector output, gas injectors have longer opening/closing time so the petrol injector pulse length needs to be offset (usually longer than petrol injector pulse especially at idle). Even if you could get around that, the offset means the gas injectors have less available time to actually be open between engine intake cycles so at high loads gas injectors need to flow a bit more than petrol injectors for a given pulse length if they are not to exceed available window (time between intake cycles, must be able to open and close within that window if pulse length isn't going to overshoot and the engine run lean even when gas injectors are open constantly with no time to close being induction cycles)... So compared to the petrol map the gas map starts at the low load area with a longer pulse length than the petrol map but ends at the high load area with around the same pulse length as the petrol map and most of the changes/compensation for this offset is done near the low load end of pulse duration. Then on top of all that, the gas system needs to compensate for things the petrol system doesn't, like gas pressure and gas temperature which effect the density of the fuel flowing through the gas injector for any given pulse length.

Even if a liquid LPG injection system is fitted the LPG system needs a different map to the petrol injection system - in this case pressure to LPG injectors needs to be higher than pressure to petrol injectors to prevent LPG turning into a vapour in feed lines to LPG injectors in warm conditions. Although it may seem liquid injection doesn't need all of the same considerations as vapour injection, in practice vapour systems are far more accomplished and likely to give the best results than liquid injection systems (for port injection and carb engines anyway).

Starting an engine on LPG - This is possible and sometimes OK with a mixer system which is a standalone fuel system that outputs gas only at atmospheric pressure. Standalone means it will ignore any 'choke' effect (cold start enrichment) the petrol system dictates (petrol needs cold start enrichment because petrol won't vaporise properly when the engine is cold, LPG is already a vapour when it enters the engine so needs no choke effect). Low pressure from a mixer reducer means there is less chance of cold gas (from a cold reducer) turning back condensing. It is possible to start the engine on LPG with other types of LPG system (besides mixer systems) fitted but is usually a bad idea due to the choke effect causing too rich mixture and due to the chance of higher pressure LPG condensing.

The only things preventing all mixer systems from being able to start well on gas is (as you implied) lack of signal at low cranking rpm, and the possibility of the reducer still getting too cold as it supplies gas to the engine (with associated cooling effect to the reducer) before the engine is providing any heat to the reducer. But it's usually possible to set a bit of 'idle bypass' at a mixer system reducer (which effectively does much the same thing as running a line from tank vapour to inlet manifold) to get decent results starting an engine on gas with a mixer system.

In very basic terms the idea of setting the power valve first is that you set the power valve somewhere near the middle then set reducer sensitivity to give correct mixture with the power valve in the middle, the power valve then gives some degree of fine tuning the mixture off idle. But the power valve has much more effect on mixture under load than at idle (when less gas has to be drawn past the power valve).. so if you then find mixture is wrong for idle you can adjust reducer sensitivity until mixture is good but this will also have a knock on effect on where the power valve needs to be set. If you can't get both good idle and good under load mixture with any combination of power valve and reducer sensitivity settings it might be because the mixer is wrong for the engine (or the reducer has @!## sensitivity). Ideally (and when a mixer is really well suited to the engine and power valve and reducer sensitivity are setup well) mixture will be near correct under all engine conditions, if a closed loop stepper power valve is then fitted the stepper won't have to move far to keep mixture correct. Then it's easy to imagine that towards stepper nearly closed conditions each stepper step will have a bigger effect on mixture than the effect of a step when the stepper is already open wider, just that each step at nearly closed position has a bigger percentage difference on the size of the restriction than the percentage difference for a step when the restriction is wider... points that can be used to set the extent of mixture swing / lambda flick. If you fit a mixer that is too small reducer sensitivity for good mixture at idle will be turned down - you could maybe fit a bit mixer (venturi) that would cause less restriction to airflow to the engine. Or if you fit a mixer that has too wide a venturi the engine might get too lean a mixture even with reducer sensitivity set to max (at least unless any idle bypass is set). I like to see mixer systems setup so mixture is near correct with little movement in stepper position regardless of engine load, but I also like to see them where at idle stepper has to be a bit more open than for under load. In the latter case it gives the impression the biggest venturi that could be used is fitted (so mixer fitted is the minimum possible restriction to airflow while giving good all round results - unless it's a poor unsensitive reducer that's fitted lol).
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

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#14 Post by Pinger » Mon Feb 24, 2020 4:43 pm

LPGC wrote:
Mon Feb 24, 2020 3:45 pm
To answer your questions thoroughly would involve a lot of typing! If you search this forum you'll find they've already been answered many times, the answers are obvious anyway when you have a decent grasp of the basics.
I did mean to mention - rather than retype, direct me to where the info is. I have been reading threads on the forum but there are so many.....
Anyway, thanks for the detailed replies. One lot lets me dismiss the (vapour phase) injection (driven by petrol ECU), the rest helps me enormously understanding the system I have.

Gilbertd
Advanced Member
Posts: 8519
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Advice please!

#15 Post by Gilbertd » Mon Feb 24, 2020 6:06 pm

and to answer your other question, why start on petrol and immediately switch to LPG rather than just start on LPG. You can. The Millennium has an alternative fuel start as an option in the setup. The difference is that when starting on petrol the fuel is injected directly into the intakes so, in my case anyway, it fires almost instantly with no cranking. As well as the lower signal, or suck, at cranking speeds, the LPG vapour is introduced before the throttle body so it has to fill the intake manifold before it gets to the cylinders so the starter has to be cranked for that much longer. There is also a danger that when the LPG first arrives the mixture may be a touch lean and could cause it to spit back into the inlet manifold and you know the affect of that!

There's nothing to stop you ditching the full size petrol tank and fitting a smaller one. I keep around a quarter tank of petrol in my car and top it up a couple of times a year unless I've had to run on petrol for any distance. A tank giving a range of 50 miles or so would be the smallest I would fit, but it can free up a lot of space.
96 Saab 900XS, AEB Leo, sold
93 Range Rover 4.2LSE, Lovato LovEco, sold
97 Range Rover 4.0SE, multipoint, sold
98 Ex-Police Range Rover 4.0, AEB Leo, daily motor
96 Range Rover 4.6HSE Ascot, AEB Leo, my spare


Proud member of the YCHJCYA2PDTHFH club.

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#16 Post by Pinger » Tue Feb 25, 2020 1:01 pm

Thanks Gilbertd.
Eliminating petrol is good for keeping engine oil clean - at its worst on cold start up....
I'm torn between retaining the petrol system (back fire prevention is one very good reason to!) at reduced capacity or retaining it at full (tank) capacity. Not sure is there is actually enough height for the LPG tank where the petrol tank currently is and even if there is, there wont be much room for even a smaller petrol tank anywhere underneath and I'd prefer to stick with Chevrolet petrol pump (and EVAP) which dictates tank height. The LPG tank indoors though deprives me of a flat floor which can otherwise take 8' x 4' sheets - spacious.
For now, I think I'll leave the petrol system as it is until I have the LPG system working as it should and then decide. Keeping petrol is also a useful safeguard for any component failure in the LPG system - worth having as breakdown insurance on a 21 year old 2500kg import may be hard to obtain.
Do you know if the AEB/Millenium system shuts down the petrol pump when running on LPG?

LPGC
Installer
Posts: 4007
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Advice please!

#17 Post by LPGC » Tue Feb 25, 2020 1:49 pm

Pinger wrote:
Tue Feb 25, 2020 1:01 pm
Do you know if the AEB/Millenium system shuts down the petrol pump when running on LPG?
You can wire it to do that but usually the petrol pump is kept running.
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

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#18 Post by Pinger » Tue Feb 25, 2020 2:21 pm

LPGC wrote:
Tue Feb 25, 2020 1:49 pm
Pinger wrote:
Tue Feb 25, 2020 1:01 pm
Do you know if the AEB/Millenium system shuts down the petrol pump when running on LPG?
You can wire it to do that but usually the petrol pump is kept running.
Those severed yellow wires?!!!

Probably worth doing. Needlessly pumping fuel at 4bar and returning it to the tank is wasteful. So much so that more modern cars now control the pump to deliver only what is required.
With the risk of fuel going stale (and that being a particular problem precisely when it is least wanted - cold start) presumably only the sealed nature of the system prevents it going 'off' when constantly churned and heated.

LPGC
Installer
Posts: 4007
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Advice please!

#19 Post by LPGC » Tue Feb 25, 2020 2:51 pm

One way of doing it is to use the yellow wires. The yellow wires, relay they're connected to in the unit and tracks on the PCB in the unit won't carry the current for the pump directly but are OK to control a relay coil (original vehicle petrol pump relay or a separate relay you fit in the petrol pump wiring). If you use the yellow wires for this purpose you've got to make sure the 'yellow wires utilisation' setting in software is set to 'disconnect petrol pump' not 'reset the ECU'.

There are pros and cons to turning off the petrol pump, if you're going to do it there are pros and cons to using the yellow wires to do it . IIrc the yellow wires are joined when the unit is switched to petrol and disconnected when running on gas, if you ever want to run on petrol with the gas ECU unit unplugged you'd have to rejoin the yellow wires, not a biggie but one of the advantages of a mixer system is in them being a completely separate fuel system and this just makes for a bit more inter-reliance between petrol and gas systems (and they can go wrong / forget settings in old age).

If you start and run on gas nearly all the time it's probably a good idea to still run the petrol pump from time to time.
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

Pinger
Junior Member
Posts: 240
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#20 Post by Pinger » Tue Feb 25, 2020 4:13 pm

Thanks LPGC - useful info.
I think, starting on petrol is going to be the way to go (avoid back fire and keep petrol pump active from time to time) then switch to LPG with the pump disabled.
I'm also going to consider how the two systems are integrated so if one fails, I can get home on the other. Hopefully I'll get back to sorting all this out soon - been held up with the weather and other things lately. Such is life!

Post Reply