Advice please!

Message
Author
Pinger
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#321 Post by Pinger » Tue Oct 27, 2020 8:30 pm

Pretty sure I enquired about that one and the ID was around that - and no larger.
260hp requires 33% more air than for 195hp. The duct I'm fitting my mixer into is 90mm ID and the throttle body is 72mm.

I need/want to better understand the gas dynamics that were present today. Clearly, my assumption that the air would recover (ie reduce in velocity and regain its pressure) after passing through the first mixer and before presenting at the second was wrong. What I created was an overall vacuum either side of the downstream mixer in addition to that created by its venturi effect - possibly greater than it's venturi effect (in essence, it was the equivalent of a good old fashioned choke). I have to ascertain if that prevails over the entire operating range and if it's even desirable. There's very possibly an element of what LPGC has eluded to re one mixer being hungrier than the other. If so, it was given ample opportunity today to make its presence known as each mixer was being fed directly from the reducer without any shared plumbing. I just need to rethink after what I've learned today and work out what next. Gas flow is a mechanical issue - governed by physics. A mechanical solution is required. I already have one in mind and it's simpler that dual mixers.

Gilbertd
Advanced Member
Posts: 8604
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Advice please!

#322 Post by Gilbertd » Tue Oct 27, 2020 10:40 pm

I take it that's 72mm ID? If 72mm OD then the Range Rover mixer I've got would slot straight on.

I'm afraid mechanics to me means bits that go up and down or round and round. Few people understand electrics as they can't see it and I'm afraid gas flow is much the same for me, I know what it does but wouldn't have a clue how to start working it out.
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
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#323 Post by Pinger » Wed Oct 28, 2020 9:56 am

Gilbertd wrote:
Tue Oct 27, 2020 10:40 pm
I take it that's 72mm ID? If 72mm OD then the Range Rover mixer I've got would slot straight on.
It wouldn't physically fit as the cowling that fits above is much larger (to accommodate the duct for the IAC valve) and would be so close to the butterfly as to create mixture distribution problems. The Vortec has its throttle body where a (single down draught) carb normally is with a V8 but offset to the front.
The duct the mixer sits in is 90mm ID.


Gilbertd wrote:
Tue Oct 27, 2020 10:40 pm
I'm afraid mechanics to me means bits that go up and down or round and round. Few people understand electrics as they can't see it and I'm afraid gas flow is much the same for me, I know what it does but wouldn't have a clue how to start working it out.
I'm beginning to gain a better understanding of what I encountered yesterday (and can expand on it later if there's interest).

I do have a question though that relates to the issue with an oversized mixer.

In essence, it is exactly the same problem as faced by carbed engines when the venturi (or choke) is oversized in pursuit of maximum power over all else. The engine will idle from its idle circuit but when the throttle is opened the idle circuit can no longer cope and there isn't sufficient air velocity in the venturi to create enough signal to draw fuel. The solution? Dip the clutch, bring the revs up enough to get the air moving and slip the clutch to keep it going. Neither viable or desirable with LPG and an auto box.
But - and I've not had the courage or wit at the time to try it - if an LPG motor attains high enough rpm to get past its immediate out of idle (tip in?) phase, will the increased air speed then allow the ('oversized') mixer to function?

LPGC
Installer
Posts: 4188
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Advice please!

#324 Post by LPGC » Wed Oct 28, 2020 10:08 am

Pinger wrote:
Tue Oct 27, 2020 8:30 pm
If so, it was given ample opportunity today to make its presence known as each mixer was being fed directly from the reducer without any shared plumbing. I just need to rethink after what I've learned today and work out what next. Gas flow is a mechanical issue - governed by physics. A mechanical solution is required. I already have one in mind and it's simpler that dual mixers.
If you feed 2 mixers from 1 reducer there's always shared plumbing, the plumbing meets at the reducer, the loop I mentioned merely becomes longer and reaches all the way back to the reducer instead of only reaching as far as the T in the pipe between mixers.

It is a physics question but a very complicated one, we can use rules of thumb but to accurately work things out such as pressures at certain points in the system / air intake through a mixer might be something a vehicle manufacturer might use a super-computer for these days and then results may need to be confirmed/proven in practice.

By all means if I was having problems like you are I might try a restriction further upstream than the point of gas entry (I think at one point I said that mixers spaced further away from the throttle body, or at least on a carb, can help airflow and signal) but I'd try to steer away from using multiple mixers plumbed to reducer(s) in series because I see potential for this further complicating things.

This model of engine has been successfully converted using a single conventional mixer mounted on the throttle body in the past.
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
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#325 Post by Pinger » Wed Oct 28, 2020 11:26 am

LPGC wrote:
Wed Oct 28, 2020 10:08 am
If you feed 2 mixers from 1 reducer there's always shared plumbing, the plumbing meets at the reducer, the loop I mentioned merely becomes longer and reaches all the way back to the reducer instead of only reaching as far as the T in the pipe between mixers.

It is a physics question but a very complicated one, we can use rules of thumb but to accurately work things out such as pressures at certain points in the system / air intake through a mixer might be something a vehicle manufacturer might use a super-computer for these days and then results may need to be confirmed/proven in practice.
Agreed - it's a ball ache I don't need.
LPGC wrote:
Wed Oct 28, 2020 10:08 am
By all means if I was having problems like you are I might try a restriction further upstream than the point of gas entry (I think at one point I said that mixers spaced further away from the throttle body, or at least on a carb, can help airflow and signal) but I'd try to steer away from using multiple mixers plumbed to reducer(s) in series because I see potential for this further complicating things.

This model of engine has been successfully converted using a single conventional mixer mounted on the throttle body in the past.
But, what yesterday did show was that a restriction upstream allows me to open up the mixer downstream (for more airflow). And said restriction is what I'd proposed as an upstream mixer. So looking to pursue that. To proceed though, it's an answer to this question:
if an LPG motor attains high enough rpm to get past its immediate out of idle (tip in?) phase, will the increased air speed then allow the ('oversized') mixer to function?
that I'm after. Any ideas? Has anyone pushed through the 'lean' barrier and found out? - and survived to tell the tale!

LPGC
Installer
Posts: 4188
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Advice please!

#326 Post by LPGC » Wed Oct 28, 2020 3:13 pm

A restriction upstream may cause partial vacuum between the restriction and mixer on the throttle body, in the right circumstances this could mean you don't need a mixer on the throttle body at all - My baked bean can 'mixer' simply had the top removed from the can to fit onto the throttle body, a hole made in the other end and a gas feed entering the side. It worked very well but I expect it would have been more limiting to max airflow than a proper mixer due to causing more turbulence.

I've seen setups that idled too lean but ran with correct rich mixture off idle. I've said before (might have been another thread) that I like to see setups have stepper motor slightly more open at idle than at off idle, it kind of shows that the venturi size (and hence restriction to intake air) limit airflow as little as possible (couldn't fit a bigger venturi or it couldn't be made to idle with rich enough mixture). Could argue that reducer idle bypass's are there to allow for this condition.

For a given size outside fitting and given size venturi 'slimline' (low profile / short) mixers are known to give less linear signal than 'full size' mixers. Again probably best to avoid trying to understand all the physics in detail (I don't claim to understand them lol) but I believe air flow will usually be fastest through the centre of a hole and if a hole is big enough a small airflow through it won't create any signal at all to a pipe connected to the side? Otherwise (and if signal was proportional to airflow regardless of venturi size) we could fit any size mixer we wanted and we'd always fit a big enough mixer as to create zero restriction.

Every venturi size will have a max airflow ability and as limits are approached will create more restriction to airflow, of course whether this restriction is enough to impact on the engine's breathing ability depends on the size of the venturi. It will also have a signal versus airflow slope that will probably start flat (no signal) for very low (relative to venturi size) airflows and probably not be fully linear versus airflow all the way through that slope. The reducer will also have a signal versus gas flow slope that will start off flat (no gas flow) for very low signal and will peak at the reducer's max flow ability given enough signal, this slope also won't be totally linear. If we plotted the 2 response slopes on a graph and added them together we'd have a 3rd slope which would show us gas flow for given airflow, high points in the 3rd slope might be rich mixture, low points in the 3rd slope might be lean mixture... Where either of the earlier 2 slopes bottomed out (lack of signal or signal response) / topped out (max signal response, max flow from the reducer) the 3rd slope would too and the engine would run with incorrect mixture at that point. Probably not even quite as simple as that though... E.g. a 5 litre engine might use 100cfm of air at full throttle at low rpm, also use 100cfm at part throttle at medium rpm, also use 100cfm at lower throttle at high rpm but in each case the mixer and signal might be effected by other dynamics besides just total airflow - assuming the mixer causes some restriction to incoming air, those 3 conditions with same airflow might see the mixer and signal be effected by different partial exposure to varying manifold vacuum.

If we had an engine that could idle very efficiently (little airflow to idle) but could generate a lot of power for it's size (could term this a wide dynamic range) we'd need the mixer venturi big enough as not to restrict airflow for the high bhp ability but that would be a big mixer venturi related to airflow for the efficient idle. In this scenario we might have a lean mixture at idle (because at idle we've fallen into the flat area on the response slopes) but correct mixture when the engine is producing more power (because we're into a more linear area on the overall response slope).
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
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#327 Post by Pinger » Wed Oct 28, 2020 5:53 pm

Pressure drop through a venturi is proportional to the square of the fluid velocity passing through it. The air requirements of an engine at WOT are pretty much linear with rising rpm - and air velocity increases are linear also.
Seems to me that the problem with larger mixers is that between idle (where the bleed can be utilised) and high rpm where the mixer may generate sufficient signal given its proportional to the square of the air speed - there's a hole to fall into just out of idle where (for me at least) lurks the backfire risk (and general unresponsiveness anyway). Carbs on petrol face the same and deployed all sorts of trickery to overcome it.

Given the mere placing of that 45mm ID mixer upstream of my current mixer permitted freeing up its area, I'm looking to deploy an upstream venturi to get through the just out of idle 'hole' while minimising the turbulence. The pressure drop of an upstream restriction (even though its ID is bigger than the feed mixer) obviously presents as a pressure drop at the mixer. But it will I think in its effect be more linear (not related to velocity squared) than the core of a venturi. The pressure drop is a consequence of that process - the friction having robbed the air of internal energy.
How restrictive this is re top end power remains to be seen but if at higher airflows the mixer steps up, then the upstream restriction can possibly be stood down. Did you ever notice that the choke plates on twin choke Webers fitted to some Fords were not symmetrical either side of their pivot shaft but could be pulled open when airflow requirements demanded?

Reducer sensitivity is a vague concept to me though. I read somewhere that it can be considered to be the equivalent of petrol height in a carb's float chamber (the higher that level, the more easily the fuel can be drawn). Is that a reasonable analogy?

Gilbertd
Advanced Member
Posts: 8604
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Advice please!

#328 Post by Gilbertd » Wed Oct 28, 2020 9:50 pm

Explained nicely here http://www.diy-lpg.co.uk/articles/files ... -loop.html but what the sensitivity adjustment does is adjust the gas pressure. Pressure is always marginally below atmospheric so has to be sucked in and the sensitivity adjustment adjusts how much for a given amount of 'suck'.
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
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#329 Post by Pinger » Thu Oct 29, 2020 10:53 am

The 'pressure head' in a carb's float chamber is a decent analogy then.

My reducer is delivering at above atmospheric pressure though (and that is with the idle bleed closed). TT say this is OK. I guess the pressure at the outlets drops pretty quickly once out of idle.

Pinger
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#330 Post by Pinger » Fri Oct 30, 2020 4:04 pm

Tried an upstream restrictor (45mm ID) with 7 of the mixer holes and it was chronically lean. At no load 3000rpm it looks like it should work (actuator at 1600. Drive it and the actuator goes fully open, it's first box green lean and that's just in the second TPS box.
Back to 7 blocked, 4 open holes on the mixer and I think that's it - for now at least.
Attachments
IMG_4932 (2).JPG
IMG_4932 (2).JPG (682.78 KiB) Viewed 2096 times
IMG_4933 (2).JPG
IMG_4933 (2).JPG (818.41 KiB) Viewed 2096 times
IMG_4934 (2).JPG
IMG_4934 (2).JPG (540.45 KiB) Viewed 2096 times

Bobby B.
New member
Posts: 2
Joined: Thu Nov 12, 2020 11:19 pm

Re: Advice please!

#331 Post by Bobby B. » Thu Nov 12, 2020 11:39 pm

Hello,
Sorry for your troubles. I have 2012 Caprice PPV 6.0 V-8 which was built by Holden in Australia which I purchased last year. It had a LPG system added to it and was a alternative fuels demo car and has only 20K miles on the odometer. The system is an eight port vapor injection type. This car runs like a champ and when put on the Dynometer it made 377HP for a 22HP gain over the 355HP on gasoline mode. I just want to let you know that there are V-8 running on propane with awesome results.

Good Luck & Best regards,
Bobby B.

Pinger
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#332 Post by Pinger » Fri Nov 13, 2020 9:18 am

Bobby B. wrote:
Thu Nov 12, 2020 11:39 pm
Hello,
Sorry for your troubles. I have 2012 Caprice PPV 6.0 V-8 which was built by Holden in Australia which I purchased last year. It had a LPG system added to it and was a alternative fuels demo car and has only 20K miles on the odometer. The system is an eight port vapor injection type. This car runs like a champ and when put on the Dynometer it made 377HP for a 22HP gain over the 355HP on gasoline mode. I just want to let you know that there are V-8 running on propane with awesome results.

Good Luck & Best regards,
Bobby B.
I'm stubbornly sticking with my simple single point system precisely because it is simple!
The more I read of the injection systems the more I appreciate how well they work. 377hp is pretty impressive. If I can get the 260hp my truck left the factory with I'll settle for that.

Pinger
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#333 Post by Pinger » Tue Dec 01, 2020 4:33 pm

Continuing a conversation from yesterday.....

What I found today was what appears to be as the AEB manual shows for O2 sensor wiring - but duplicated.
From the O2 sensor, that wire has been cut and connected to a purple wire that goes to the AEB ECU. From the AEB ECU a grey wire goes to the GM ECU (where the wire from the O2 sensor would have originally gone before being cut). Presumably all this is as expected.

There is also exactly the same on a second wire (presumably from the other bank O2 sensor) the only difference is that the wires coloured as above have a black streak.
The AFR gauge I'm trying to wire when hooked up to a battery shows a voltage (0.7V falling to 0.2V) between its sensor signal wire (the one I have to connect to an O2 sensor) and ground so better not to parallel it with any of the above?

Should I undo one of the O2 sensors ie, cut a grey wire, reconnect the O2 sensor wire to the GM (as it originally was from the factory) and just parallel the gauge into that?
Or, just parallel it with one of the two sensors and ignore any bias(?) voltage as this is how it's recommended it is wired and, I'll be doing the same as I'm avoiding on the AEB ECU to the GM ECU - which will likely throw the EML on if I don't have it connected to its ECU? If I isolate an O2 sensor, will the AEB system run as before without it - despite being configured (it would seem) for two?

Gilbertd
Advanced Member
Posts: 8604
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Advice please!

#334 Post by Gilbertd » Tue Dec 01, 2020 6:02 pm

Huh? Some of the multipoint systems for 5, 6 and 8 cylinder engines have a duplicate pair of lambda sensor wires so you get a pair with Purple and Grey and another pair with Purple/Black and Grey/Black but you wouldn't get this on a Leonardo as it is a controller for a single point system so there's no point in comparing lambda from two banks when the fuel only goes in from one place. Think about it, what would the actuator do if one lambda was showing weak and the other showing rich? Rather than connect as per the AEB diagram and cutting the lambda signal wire, leave it connected through to the GM ECU and Tee the Purple wire onto that. insulate the end of the Grey wire and leave it disconnected. For the AFR gauge, just Tee the sensor signal wire into the lambda sensor signal wire. If you cut the lambda wire, the GM ECU will see that as a low voltage (meaning a weak mixture) so will, at best, conclude the sensor has died and go into a default fuelling strategy or, at worst, wind the mixture richer and richer until that one bank is drowning in petrol.
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
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#335 Post by Pinger » Tue Dec 01, 2020 6:44 pm

Gilbertd wrote:
Tue Dec 01, 2020 6:02 pm
Huh? Some of the multipoint systems for 5, 6 and 8 cylinder engines have a duplicate pair of lambda sensor wires so you get a pair with Purple and Grey and another pair with Purple/Black and Grey/Black but you wouldn't get this on a Leonardo as it is a controller for a single point system so there's no point in comparing lambda from two banks when the fuel only goes in from one place. Think about it, what would the actuator do if one lambda was showing weak and the other showing rich?
Absolutely - the system couldn't respond to differing signals but definitely there are the duplicated wires. One pair solid colours and the other pair the same colours with the black stripes.
What I went looking for was the O2 sensor connected to the AEB so that I could use the other for the gauge. Have I got some kind of mutant AEB system? It's branded Millenium and my understanding was that it and the Leonardo are pretty much the same thing.

Gilbertd wrote:
Tue Dec 01, 2020 6:02 pm
Rather than connect as per the AEB diagram and cutting the lambda signal wire, leave it connected through to the GM ECU and Tee the Purple wire onto that. insulate the end of the Grey wire and leave it disconnected. For the AFR gauge, just Tee the sensor signal wire into the lambda sensor signal wire. If you cut the lambda wire, the GM ECU will see that as a low voltage (meaning a weak mixture) so will, at best, conclude the sensor has died and go into a default fuelling strategy or, at worst, wind the mixture richer and richer until that one bank is drowning in petrol.
So, what I need to do is reconnect as per GM's wiring, tee one of the sensors to the AEB and the other to the AFR gauge?

Gilbertd
Advanced Member
Posts: 8604
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Advice please!

#336 Post by Gilbertd » Tue Dec 01, 2020 7:03 pm

It'll be branded Millennium and will have Millennium firmware but if you look at the circuit board, it's still an AEB175, so quite what the other pair of wires are doing is anyone's guess. Can you pull the boot on the plug back and see where the extra pair of wires are connected?

Yes, reconnect as per GM and Tee into the two signal wires with the AEB Purple and signal wire for the AFR gauge.
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
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#337 Post by Pinger » Tue Dec 01, 2020 8:13 pm

Gilbertd wrote:
Tue Dec 01, 2020 7:03 pm
It'll be branded Millennium and will have Millennium firmware but if you look at the circuit board, it's still an AEB175, so quite what the other pair of wires are doing is anyone's guess. Can you pull the boot on the plug back and see where the extra pair of wires are connected?
Weather and time permitting I'll try and see if they remain separate.
Weird thing is it follows the GM convention of same colour wires but with additional stripe to delineate between banks. According to Haynes at least - my truck has four pairs all the same colouring from all four O2 sensors. Working out which is which is nigh on impossible. When I re-jig them tomorrow I'll stick with how they've been aligned so far in the hope they haven't been muddled. Won't matter for LPG but for petrol it may.
Gilbertd wrote:
Tue Dec 01, 2020 7:03 pm
Yes, reconnect as per GM and Tee into the two signal wires with the AEB Purple and signal wire for the AFR gauge.
Hopefully I'll get that done tomorrow. Just curious that wired so differently the AEB ECU won't notice any difference or operate differently?

Gilbertd
Advanced Member
Posts: 8604
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Advice please!

#338 Post by Gilbertd » Wed Dec 02, 2020 1:54 am

It won't affect the AEB as it will still be seeing the same signal as it was before. Difference is that the lambda sensor signal to the GM ECU will be going direct and not via the AEB controller. It should just pass it through but can sometimes mess with it and send a simulated signal (or you can make it send a simulated signal) which can screw up running on petrol.
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.

LPGC
Installer
Posts: 4188
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Advice please!

#339 Post by LPGC » Wed Dec 02, 2020 11:52 am

Connecting the grey wire can be useful on some installs, that's why it's there lol.

With a closed loop mixer system fitted on a closed loop petrol injection engine there'll always be a difference in how the 2 closed loop systems react to mixture readings. We know that if we drive the car on LPG for long enough eventually the petrol system fuel trims will drift to either extremely lean or extremely rich... and perhaps lean at some engine loads (rpm and throttle combinations) while rich at other engine loads.

If we don't use the grey wire we leave it to chance whether the petrol trims will go lean or rich. The voltage on the grey wire can be programmed to fluctuate as we want, so we can program the grey wire signal to steer petrol trims to go lean or rich. Some vehicles use fuel trim info even during engine starting and/or full throttle (during high load mixture enrichment) so can have starting and/or driveability problems if fuel trims go extreme lean or rich... but usually less problems if trims go one way rather than the other.

On most early fuel injected cars it doesn't really matter, the engine will start and run OK on petrol even if trims are extreme rich or extreme lean. But even some of these will turn on the Mil light if they don't see the lambda signal voltage flick between a lean and rich reading regularly enough. A closed loop mixer system should see the lambda reading flick regularly anyway but using the grey wire can make certain that it sees the flick. The mixer system may hold lambda voltage above 0.5v above 4000rpm, some petrol systems may turn on the Mil light under those conditions unless they see lambda voltage flick.

Ultimately the grey wire has limited uses on a small number of vehicles and even on those vehicles (and if keeping fuel trims something like neutral makes for better starting on petrol etc) it's better to fit something like an AEB426 which connects to OBD to read fuel trims and provides it's own emulated lambda signal which is modulated to keep fuel trims at zero. But we can't use the AEB42X range on every vehicle with a mixer system because they're only available to work with certain OBD protocols and certain types of lambda sensor.

Example: The pre 2000 4Litre Jeeps could run happily on a mixer system even without using the grey wire but using the grey wire could help steer trims to make for better engine starting. Fit the same basic mixer system onto a post 2000 4litre Jeep and it would still run happily on gas but the Mil would come on and the cruise control would be disabled. Connect the grey wire and it could run for a much longer time before the Mil came on but it would inevitably still come on at some point. Wire in an AEB426 and all the problems went away.
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
Member
Posts: 284
Joined: Tue Feb 11, 2020 5:08 pm

Re: Advice please!

#340 Post by Pinger » Wed Dec 02, 2020 1:17 pm

LPGC wrote:
Wed Dec 02, 2020 11:52 am
Connecting the grey wire can be useful on some installs, that's why it's there lol.

With a closed loop mixer system fitted on a closed loop petrol injection engine there'll always be a difference in how the 2 closed loop systems react to mixture readings. We know that if we drive the car on LPG for long enough eventually the petrol system fuel trims will drift to either extremely lean or extremely rich... and perhaps lean at some engine loads (rpm and throttle combinations) while rich at other engine loads.

If we don't use the grey wire we leave it to chance whether the petrol trims will go lean or rich. The voltage on the grey wire can be programmed to fluctuate as we want, so we can program the grey wire signal to steer petrol trims to go lean or rich. Some vehicles use fuel trim info even during engine starting and/or full throttle (during high load mixture enrichment) so can have starting and/or driveability problems if fuel trims go extreme lean or rich... but usually less problems if trims go one way rather than the other.

On most early fuel injected cars it doesn't really matter, the engine will start and run OK on petrol even if trims are extreme rich or extreme lean. But even some of these will turn on the Mil light if they don't see the lambda signal voltage flick between a lean and rich reading regularly enough. A closed loop mixer system should see the lambda reading flick regularly anyway but using the grey wire can make certain that it sees the flick. The mixer system may hold lambda voltage above 0.5v above 4000rpm, some petrol systems may turn on the Mil light under those conditions unless they see lambda voltage flick.

Ultimately the grey wire has limited uses on a small number of vehicles and even on those vehicles (and if keeping fuel trims something like neutral makes for better starting on petrol etc) it's better to fit something like an AEB426 which connects to OBD to read fuel trims and provides it's own emulated lambda signal which is modulated to keep fuel trims at zero. But we can't use the AEB42X range on every vehicle with a mixer system because they're only available to work with certain OBD protocols and certain types of lambda sensor.

Example: The pre 2000 4Litre Jeeps could run happily on a mixer system even without using the grey wire but using the grey wire could help steer trims to make for better engine starting. Fit the same basic mixer system onto a post 2000 4litre Jeep and it would still run happily on gas but the Mil would come on and the cruise control would be disabled. Connect the grey wire and it could run for a much longer time before the Mil came on but it would inevitably still come on at some point. Wire in an AEB426 and all the problems went away.
As far as I can ascertain my petrol fuel trims haven't deviated to the point of being problematic (it starts promptly on petrol) and I don't want to lose that.
If utilisation of the grey wire(s) is helping this should I consider just tagging the AFR meter to one of the O2 sensor wires?

This:''The mixer system may hold lambda voltage above 0.5v above 4000rpm,'' is a feature of mine. If not the 4000 rpm threshold then a (configurable) TPS threshold. That is one of the things I want to monitor (and tweak) with the AFR meter.

Post Reply