I spent this weekend finishing the wiring and making a mount for the coil packs.
The engine started perfecting first turn of the key. When the engine is warm, I have found that i get quite of a log of backfires on LPG when cranking. This has led to a few blown off hoses.
I set the cranking advance to 8 degrees. Which hasn't really stopped the backfires on start up, but it appeared to help. However, this morning I found the car was extremely hard to start, with the crank advance of 8 degrees.
I thought the backfire might be due to too much advance at startup, but it seems that 8 degrees might be too low for a cold start.
I still have the stock coil wired up, with the HT plugged to stop it arcing to the body. This means that I still have the ECU feed coming from the negative of the stock coil. The car seems to run quite well on petrol in this state. And when started it run well and drives well on LPG. Just the backfire when trying to start, that plagues me, as a blown off hose stop the car from starting.
I also built the diode and zener diode circuit for the feed from the coil packs, and i tried running the ECU feed from this, but I found the car ran really rough. I ran the feed from the zener circuit through the stock resistor. Should I leave the stock resistor in place, or am i bringing the feed down too low for the ECU to register it.
In short, what sort of cranking advance should I ran? What is ideal to help cold starts and stop backfires on start up. Or is something else causing backfire at startup? And, do i need the stock resistor in place when running the through the zener diode?
Does anyone have some advice?
Thanks in advance for any help.
John.
89 RRC 3.5 LPG. New MJ install, cold start and backfires.
Moderators: JeffC, rdoherty, stieg, brentp
Tonight after work, I thought I would try starting the car cold. After 5 or 6 serparate tries at cranking, it would not fire. The cranking advance is set to in MJ.
I then removed the MJ and tried with EDIS only. The car started second try,and sat at a rock solid 10 degrees. Or maybe 9, it is a bit hard to see the timing marks, but I might be a degree out.
What could be causing the backfire on startup?
It might be a coincedence, but it seems to cold start better on EDIS only.
Could being 1 or 2 degrees out with the trigger wheel be the problem. If I am sparking at 9, should I set the offset to -1 or +1.
Could this mean cranking advance or 8 is actually 7, which is getting too low for starting on LPG.
Any words of wisdom would be helpful.
John.
I then removed the MJ and tried with EDIS only. The car started second try,and sat at a rock solid 10 degrees. Or maybe 9, it is a bit hard to see the timing marks, but I might be a degree out.
What could be causing the backfire on startup?
It might be a coincedence, but it seems to cold start better on EDIS only.
Could being 1 or 2 degrees out with the trigger wheel be the problem. If I am sparking at 9, should I set the offset to -1 or +1.
Could this mean cranking advance or 8 is actually 7, which is getting too low for starting on LPG.
Any words of wisdom would be helpful.
John.
I have a 3.9 EFI Range Rover with MJLJ and dual fuel.
I usually start on petrol and then switch to gas, but I have my cranking advance for LPG set to 10deg and have experienced no problems so far. My maps are nowhere near optimised, but appear to work ok so I will enclose them here for you to look at. (I don't want to load them into the Maps section since they are still quite crude, but then they do work amazingly well!)
On the wiring front, I use the 'four diodes and a zenner' wiring and then put the output straight into what was the coil terminal so yes - it works fine with the current limiting resistor in place. One thing that I might mention though is that I have used shielded cable for this part of the circuit since we are talking transient spikes and things and I thought this might be influenced by induction from other noisy sources...
I usually start on petrol and then switch to gas, but I have my cranking advance for LPG set to 10deg and have experienced no problems so far. My maps are nowhere near optimised, but appear to work ok so I will enclose them here for you to look at. (I don't want to load them into the Maps section since they are still quite crude, but then they do work amazingly well!)
On the wiring front, I use the 'four diodes and a zenner' wiring and then put the output straight into what was the coil terminal so yes - it works fine with the current limiting resistor in place. One thing that I might mention though is that I have used shielded cable for this part of the circuit since we are talking transient spikes and things and I thought this might be influenced by induction from other noisy sources...
- Attachments
-
- LPG.mjlj
- (719 Bytes) Downloaded 628 times
-
- Petrol.mjlj
- (718 Bytes) Downloaded 560 times
Hello.
Thanks a lot for the help. The maps will be very helpful.
I have always been able to start on LPG, with only a very rare backfire. With the EDIS and MJ it seems every time I crank I will get a backfire.
I am using second hand Ford leads, I will swap for new ones to rule out the leads.
I will also try starting on petrol and switching to gas. I will looking into getting a changeover switch with petrol start.
Thanks again...
John.
Thanks a lot for the help. The maps will be very helpful.
I have always been able to start on LPG, with only a very rare backfire. With the EDIS and MJ it seems every time I crank I will get a backfire.
I am using second hand Ford leads, I will swap for new ones to rule out the leads.
I will also try starting on petrol and switching to gas. I will looking into getting a changeover switch with petrol start.
Thanks again...
John.
Hello All.
I found that my trigger wheel was 2 degrees out, so I adjusted for this in the MJ. Then I set the cranking advance to 10.
The static of the dizzy was set to 8, and it used to start on gas 99.99 of the time without backfire. On EDIS and MJ, it just doesn't want to start when cold on lpg. I find this a little odd.
Now I have adjusted the offset in the MJ, it only occasionally backfires when starting warm on gas.
The engine idles and revs beautifully, and pulls nicely through the rev range.
I found that with the 4xDiode, 1xZener circuit, the engine ran really rough and sounded like it was missing on petrol. I put a multimeter to the zener output and I was only getting 8 VDC. This sounds too low to me. Did I get the wrong zener? I asked to a 14V, shouldn't I get something like that?
I tested the voltage before the zener and it was reading about 20VDC. Having read that the inline resister already on the ECU input should protect the ECU, i thought I would bite the bullet and try it running the ecu driver without the zener. The car runs nicely on petrol now, all be it it rich, but it always did. I believe my flapper AFM is shagged, which it always was, and probably more so now, after the many backfires I have had.
I have to tidy up the ign lead routing quite a bit. I have to shorten a number of the leads and I do have several cross overs. I relise this could be a contributor to the backfires.
I guess my questions are:
Why would I now find it harder to start cold on gas, then I did with the dizzy?
What voltage should I be feeding the ECU from the zener output?
I am in dangerous territory feeding the ECU without the zener? Is the 20VDC I am seeing at this point too high, or will the resistor be doing the trick?
With the confidence I have gained from playing with the MJ, and a most likely shagged flapper AFM. I am starting to consider going megasquirt
In the mean time, anyone got a ok condition flapper ECU that want to part with....
Rdgs,
John.
I found that my trigger wheel was 2 degrees out, so I adjusted for this in the MJ. Then I set the cranking advance to 10.
The static of the dizzy was set to 8, and it used to start on gas 99.99 of the time without backfire. On EDIS and MJ, it just doesn't want to start when cold on lpg. I find this a little odd.
Now I have adjusted the offset in the MJ, it only occasionally backfires when starting warm on gas.
The engine idles and revs beautifully, and pulls nicely through the rev range.
I found that with the 4xDiode, 1xZener circuit, the engine ran really rough and sounded like it was missing on petrol. I put a multimeter to the zener output and I was only getting 8 VDC. This sounds too low to me. Did I get the wrong zener? I asked to a 14V, shouldn't I get something like that?
I tested the voltage before the zener and it was reading about 20VDC. Having read that the inline resister already on the ECU input should protect the ECU, i thought I would bite the bullet and try it running the ecu driver without the zener. The car runs nicely on petrol now, all be it it rich, but it always did. I believe my flapper AFM is shagged, which it always was, and probably more so now, after the many backfires I have had.
I have to tidy up the ign lead routing quite a bit. I have to shorten a number of the leads and I do have several cross overs. I relise this could be a contributor to the backfires.
I guess my questions are:
Why would I now find it harder to start cold on gas, then I did with the dizzy?
What voltage should I be feeding the ECU from the zener output?
I am in dangerous territory feeding the ECU without the zener? Is the 20VDC I am seeing at this point too high, or will the resistor be doing the trick?
With the confidence I have gained from playing with the MJ, and a most likely shagged flapper AFM. I am starting to consider going megasquirt
In the mean time, anyone got a ok condition flapper ECU that want to part with....
Rdgs,
John.
Why would I now find it harder to start cold on gas, then I did with the dizzy?
Well you should have a much better and steadier spark so - assuming the timing is ok - could it be that the gas control unit isn't giving enough? With mine I start on petrol with no throttle (I let the ECU take care of that) but if I start on gas I have to give it some on the pedal to get the (simple draw-through) gas system to see enough vacuum and start supplying gas. Then again if it was working ok before and you haven't changed any gas setting it can't be that. No - this has me stumped. Incidentally I use only 6deg for cranking on gas and get no backfires.
What voltage should I be feeding the ECU from the zener output?
Re the voltage - it is transient ans so cannot be accurately measured using a multimeter. If you imagine without the zenner a spiky trace going up to many volts and your meter averages (by some undefined means) this to 20v. Then you put in the zenner which clips the high volts, but does nothing to the rest of the signal which is within it's limit - the new 'average' may well be 8 even though the zenner is a 14v unit. Re feeding the ECU - as long as you still have the original land-rover current-limiting resistor in the circuit then you should be ok. So have you now removed the original coil from the circuit? I only ask because if this is trying to fire but not managing it then it will be experiencing huge internal voltages I would imagine and probably won't be too happy about it.
I am in dangerous territory feeding the ECU without the zener? Is the 20VDC I am seeing at this point too high, or will the resistor be doing the trick?
The (original Land Rover) resistor will do the trick just fine.
With the confidence I have gained from playing with the MJ, and a most likely shagged flapper AFM. I am starting to consider going megasquirt
You going to let MS control the gas then? If not then a replacement flapper (or even a move to hotwire) might be simpler/cheaper, after all you will be running on gas most of the time, yes?. Have a look at the MS tuning threads on LR4x4.com - everyone says how easy it is to set up petrol on MS, but there are hundreds of questions...
Well you should have a much better and steadier spark so - assuming the timing is ok - could it be that the gas control unit isn't giving enough? With mine I start on petrol with no throttle (I let the ECU take care of that) but if I start on gas I have to give it some on the pedal to get the (simple draw-through) gas system to see enough vacuum and start supplying gas. Then again if it was working ok before and you haven't changed any gas setting it can't be that. No - this has me stumped. Incidentally I use only 6deg for cranking on gas and get no backfires.
What voltage should I be feeding the ECU from the zener output?
Re the voltage - it is transient ans so cannot be accurately measured using a multimeter. If you imagine without the zenner a spiky trace going up to many volts and your meter averages (by some undefined means) this to 20v. Then you put in the zenner which clips the high volts, but does nothing to the rest of the signal which is within it's limit - the new 'average' may well be 8 even though the zenner is a 14v unit. Re feeding the ECU - as long as you still have the original land-rover current-limiting resistor in the circuit then you should be ok. So have you now removed the original coil from the circuit? I only ask because if this is trying to fire but not managing it then it will be experiencing huge internal voltages I would imagine and probably won't be too happy about it.
I am in dangerous territory feeding the ECU without the zener? Is the 20VDC I am seeing at this point too high, or will the resistor be doing the trick?
The (original Land Rover) resistor will do the trick just fine.
With the confidence I have gained from playing with the MJ, and a most likely shagged flapper AFM. I am starting to consider going megasquirt
You going to let MS control the gas then? If not then a replacement flapper (or even a move to hotwire) might be simpler/cheaper, after all you will be running on gas most of the time, yes?. Have a look at the MS tuning threads on LR4x4.com - everyone says how easy it is to set up petrol on MS, but there are hundreds of questions...
I was thinking the same thing, that maybe the backfire is due to wasted spark. I know my gas set up is running extremely rich at the moment, so on startup I was thinking unburnt gas it being ignited on the exhaust stroke of the wasted spark cylinder.. Just a thought...
I have shelved the MJ for now, as the backfired issue plagued me... My dizzy setup was more predictable.
I have shelved the MJ for now, as the backfired issue plagued me... My dizzy setup was more predictable.