Gas pressure issues

[rossko]

With no pressure - it will go "beep, tank empty"

Maybe the reducer's vacuum port is blocked internally or has a bad leak.

[LPGC]

MAP readings look fine.

Pressure readings look barmy - its as though the software thinks it is looking at an absolute type, but is really double-adjusting the reading from a differential type.
You're really really sure you have selected the correct type in the software? What type is it?

I can see the reason behind those thoughts but I think to confirm that we'd need to see the figures while the vehicle is driven... The hardest we see the engine work is about 2:30 into the video but that is only revving the engine lightly with no load - even then map doesn't go much above .4, Pinj doesn't go much above 5ms.

I think the software readings do lag a bit behind actual map/pressure readings with Nevo, the Tomasetto reducer probably briefly over-swings on relative gas pressure when manifold pressure increases (seems to be a design thing and not really a bad one). I have seen similar readings with KME / Tomasetto setups. In addition, and this is probably a bigger point, I might expect this reducer not to perform properly with the pressure set as low as it is (.88 bar). Too low a pressure setting and there will probably be inconsistent spring pressure on the diaghpram. The combination of the above might make for those strange figures we see on screen... I would increase the relative pressure to at least .95 bar with the engine running on gas and see what happens then. I bet if you attempt to increase the pressure you'll find that the allen adjuster is almost at it's extreme low setting and initially feels slack (because there is no loading on the spring), but then turning it a bit further you'll start to feel it get a little tighter as you begin to load the spring against the diaghpram. Might have been better to go a nozzle size smaller and turn the reducer pressure up quite a lot but I would expect a pressure of .95 to get the reducer into it's proper working range.

Simon

[luke123]

So today we visited Yorkshire this morning so i thought perfect time for a decent mapping session on the mours etc.

So the video below i mapped it but the EML came on during the journey home, i plugged it in and came back as system too lean bank 1 so cleared the fault code, whent for a drive upto preston to drop the brothers girlfriend off so another mapping session it was ive mapped it better i will put a pic up tomorrow showing the model on the map as it looks alot better than it did before.

Iv'e contacted LPG Shop on friday and they should be sending me another map/pressure sensor out as i think it's that at fault as the vapouriser is working as the manifold pressure goes more it's operating the diaphragm and it's been spiking to 1.6 bar of pressure.

https://www.youtube.com/watch?v=s8YLtnX ... NIGI-qpmcQ

[LPGC]

Looks like it's doing what Rosko said... Your video shows the ECU reading increasing relative gas pressure as engine manifold pressure rises but this is very unlikely to be actually happening. Most likely your reducer will be correctly keeping gas pressure constant relative to manifold pressure, even though your ECU is reading increasing relative gas pressure as manifold pressure rises. If the gas pressure is to keep relative to manifold pressure then obviously the gas pressure will need to rise as manifold pressure rises, but only by the same amount as the manifold pressure. The ECU needs to read the relative pressure but instead it seems to be reading the absolute pressure. If this is the case then as manifold pressure increases so will the gas pressure reading. When the ECU reads an over-pressure condition it will decrease the Ginj to compensate for the higher pressure. Your map looks like it is set up to compensate for the ECU reading untrue higher pressures at higher engine loads (your map starts with Ginj's that are lower than Pinj's at the low load end and has much higher Ginj's than Pinj's at the high load end). This is just what we'd expect if your ECU was reading absolute pressure instead of relative pressure...

It is possible that your map sensor is at fault (the gas pressure side of it anyway). If the map sensor were to read, say, double the actual gas pressure, then this sort of situation could occur. It doesn't look like this scenario is exactly what is happening in your case, because gas pressure seemed to read correct when you were running on petrol, so I wonder if there is an error in the settings you have made in the software...

You haven't included a screen shot of the 'automatic settings' screen in the Nevo software. Have you accidentally selected the 'Valvetronic' option? Or maybe forgot to connect the vacuum connection before you first started the calibration (If you went through the 'automatic settings' procedure, the software may have selected Valvetronic automatically if you made this mistake)? There are a lot of options / screens etc in Nevo - Would advise resetting the ECU (factory settings button), starting the calibration afresh and see how it goes.

Simon

[luke123]

Hi LPGC,

I will screen shot it tonight and post it up, i selected valve tronic as the engine is a valve tronic engine? Or are you meant to select this option after the mapping?

Cheers

Luke

[Gilbertd]

i selected valve tronic as the engine is a valve tronic engine?

Are you sure? I've just had a quick Google and it seems it has variable valve timing but not lift so isn't valvetronic.

[luke123]

i selected valve tronic as the engine is a valve tronic engine?

Are you sure? I've just had a quick Google and it seems it has variable valve timing but not lift so isn't valvetronic.

I shall have another go then tonight with out the valve tronic option and see if that changes it. All the vacuum pipes are on and plumbed in, i selected valve tronic as i thought it would be valve tronic.

[Gilbertd]

What you've got is variable valve timing, known as VVTi, VVC, etc depending on what the manufacturer decided to call it. What this does is alter the intake cam timing so at low revs you get better tractability and economy but at higher revs you get more power. A good example is the 1.8 VVTi engine used in numerous Toyota models (which I have become very familiar with recently). Below 4,000 rpm is is very smooth and tractable and allows them to quote 40+ mpg figures. Above 4,000 rpm it goes like stink but you can watch the fuel gauge drop.

Valvetronic also alters the amount of valve lift so the valves only open by a small amount at low revs and a larger amount at higher revs. The valve lift affects the revs far more than the throttle butterfly which does not a lot at all.

[LPGC]

Once you have eliminated the impossible, whatever remains, however improbable, must be the truth.

Hence my reasoning for the Valvetronic setting

Like Gilbert has said, your engine is not Valvetronics. Valvetronic models do have a throttle but the point of Valvetronic is to replace the function of the throttle (except for when the engine is warming or when there is a problem with the Valvetronic system). When Valvetronic is operational the throttle is usually fully open (meaning there won't be any manifold vacuum), and Valvetronic controls airflow by adjusting valve lift and open duration. The advantage of Valvetronic is that it can emulate an almost infinite range of camshafts - some cam durations would be better for low rpm economy, some for low rpm torque, ditto mid range / high rpms and all the points in-between. When you want more power from the engine the valve lift increases to allow more air into the engine. The ability to quickly adjust cam lift makes for better throttle response than a throttle controlled engine. In addition, one of the in-inefficiencies of a normal throttle controlled engine is that the engine has to suck against manifold vacuum, with Valvetronic there is no manifold vacuum. We can see that your engine isn't running Valvetronic because the manifold pressure goes up and down as you drive, and we know it isn't a Valvetronic engine anyway because Suzuki don't make Valvetronic engines!

Different LPG ECU's seem to have different takes /methodologies on how to deal with Valvetronic engines. Seems the Nevo might just totally disregard manifold and relative gas pressure to use only absolute pressure as a pressure reference. There are various implications to consider when converting any valvetronic engine. These won't affect you because your engine isn't Valvetronic and you shouldn't have the Valvetronic option selected but I wrote some of the implications here: http://www.lpgforum.co.uk/viewtopic.php?f=28&t=13504

Like I said above, would do a complete reset of the ECU and start again, but don't set the Valvetronic option.

Simon

[luke123]

I adjusted the software and recalibrated unclicking the valve tronic option, gas pressure Is ok now running 1.4 and software Is set 1.3.

Still getting the problem where If accelerated to quickly or changing gear and coming of the throttle to quickly It will still knock back to petrol,I also loose connection at the laptop aswell but I think thats something to do with using It via usb port.
The petrol injection timings when running on gas have dropping to 2.4m/s. So I was thinking should I drill the nozzles out, which I ended up doing and drilling them to 3mm but I think it's slightly bigger. The engine In the ignis Is the high compression one think It's 110bhp.
So I did all this and recalibrated It etc but still does the same thing just not as much.

Can this also happen If there Is too much gas being injected when accelerating, coming of the throttle and changing gear?

Iv'e set the map up to run rich as If I would of left the map where it was without adding gas i get a lean fault code. I have left the map with a 10% add accross the map but the mil light came on and read as a lean condition, so i think more gas is needed to be added on the map adjustments.

cheers Luke

[LPGC]

I adjusted the software and recalibrated unclicking the valve tronic option, gas pressure Is ok now running 1.4 and software Is set 1.3.

Good, thought that would sort the apparent pressure problem out.

Still getting the problem where If accelerated to quickly or changing gear and coming of the throttle to quickly It will still knock back to petrol,I also loose connection at the laptop aswell but I think thats something to do with using It via usb port.

On KME systems it is possible to have the system detect an error due to over-pressure, most systems don't have that facility, make sure this isn't enabled... Bare in mind it will also switch back to petrol if rpm falls low enough (again check settings) - if rpm falls too low when you dip the clutch this may be due to bad calibration at low loads. I think it more likely it switches back to petrol when you get back on the throttle due to detecting low gas pressure (thinking you have run out of fuel)... This could be due to settings you have made or because the reducer cannot physically flow as much gas as you have set in the calibration (unlikely on a 100bhp engine with Arctic reducer but what pressure have you set for changing back to petrol?).. You may be right about your interface being the problem when losing connection but this would be unusual with a KME system, so worth checking the connection loss and switching back to petrol are not associated - the system needs more power to keep injectors open for longer at higher loads and rpms - are your ECU's constant live and earth power connections good enough?

he petrol injection timings when running on gas have dropping to 2.4m/s. So I was thinking should I drill the nozzles out, which I ended up doing and drilling them to 3mm but I think it's slightly bigger. The engine In the ignis Is the high compression one think It's 110bhp.

Drilling injectors out when Pinj is lower than you want is counter-intuitive but whether or not this was a bad decision will depend on your map... If you had fitted better flowing or more linear injectors (most injectors )with 3mm nozzles on your model of vehicle you would now have a problem with too large nozzles but you should just about get away with 3mm nozzles with the V30 injectors.. though you may need to decrease pressure. Can you post another pic showing your new map?

Can this also happen If there Is too much gas being injected when accelerating, coming of the throttle and changing gear?

If you read into the implications of my answers above in this post I have answered this question - If your reducer was borderline and you set mixture too rich at the top end, then yes, but your reducer should easily cope with a 100bhp engine with loads of available flow in reserve, so no. Thus either you have a bad fuel feed to your reducer (again unlikely) or the problems are due to your software settings / vapour pipe runs / vacuum pipe connections especially to the reducer - If your reducer doesn't see an immediate rise in manifold pressure when the map sensor does then it will not 'know' to increase it's pressure output to keep gas pressure relative above manifold pressure.

v'e set the map up to run rich as If I would of left the map where it was without adding gas i get a lean fault code. I have left the map with a 10% add accross the map but the mil light came on and read as a lean condition, so i think more gas is needed to be added on the map adjustments.

I wouldn't do that, should be no need and this complicates matters from a diagnostic / calibration point of view. Another couple of reasons why the system might change back to petrol when you put your foot down - you could have set a maximum engine load on gas / max rpm on gas / petrol injectors full open on gas / gas injectors full open on gas criteria for the ECU to switch back to petrol.

Now you have re calibrated with 3mm nozzles we need to see your 'model' again. Might be helpful to clear gathered plotted points in the 'map' screen, drive on petrol and let it regather plotted points, drive on gas and let it gather plotted points - and repost a pic of this screen too.

Simon

[luke123]

I adjusted the software and recalibrated unclicking the valve tronic option, gas pressure Is ok now running 1.4 and software Is set 1.3.

Good, thought that would sort the apparent pressure problem out.

Still getting the problem where If accelerated to quickly or changing gear and coming of the throttle to quickly It will still knock back to petrol,I also loose connection at the laptop aswell but I think thats something to do with using It via usb port.

On KME systems it is possible to have the system detect an error due to over-pressure, most systems don't have that facility, make sure this isn't enabled... Bare in mind it will also switch back to petrol if rpm falls low enough (again check settings) - if rpm falls too low when you dip the clutch this may be due to bad calibration at low loads. I think it more likely it switches back to petrol when you get back on the throttle due to detecting low gas pressure (thinking you have run out of fuel)... This could be due to settings you have made or because the reducer cannot physically flow as much gas as you have set in the calibration (unlikely on a 100bhp engine with Arctic reducer but what pressure have you set for changing back to petrol?).. You may be right about your interface being the problem when losing connection but this would be unusual with a KME system, so worth checking the connection loss and switching back to petrol are not associated - the system needs more power to keep injectors open for longer at higher loads and rpms - are your ECU's constant live and earth power connections good enough?

he petrol injection timings when running on gas have dropping to 2.4m/s. So I was thinking should I drill the nozzles out, which I ended up doing and drilling them to 3mm but I think it's slightly bigger. The engine In the ignis Is the high compression one think It's 110bhp.

Drilling injectors out when Pinj is lower than you want is counter-intuitive but whether or not this was a bad decision will depend on your map... If you had fitted better flowing or more linear injectors (most injectors )with 3mm nozzles on your model of vehicle you would now have a problem with too large nozzles but you should just about get away with 3mm nozzles with the V30 injectors.. though you may need to decrease pressure. Can you post another pic showing your new map?

Can this also happen If there Is too much gas being injected when accelerating, coming of the throttle and changing gear?

If you read into the implications of my answers above in this post I have answered this question - If your reducer was borderline and you set mixture too rich at the top end, then yes, but your reducer should easily cope with a 100bhp engine with loads of available flow in reserve, so no. Thus either you have a bad fuel feed to your reducer (again unlikely) or the problems are due to your software settings / vapour pipe runs / vacuum pipe connections especially to the reducer - If your reducer doesn't see an immediate rise in manifold pressure when the map sensor does then it will not 'know' to increase it's pressure output to keep gas pressure relative above manifold pressure.

v'e set the map up to run rich as If I would of left the map where it was without adding gas i get a lean fault code. I have left the map with a 10% add accross the map but the mil light came on and read as a lean condition, so i think more gas is needed to be added on the map adjustments.

I wouldn't do that, should be no need and this complicates matters from a diagnostic / calibration point of view. Another couple of reasons why the system might change back to petrol when you put your foot down - you could have set a maximum engine load on gas / max rpm on gas / petrol injectors full open on gas / gas injectors full open on gas criteria for the ECU to switch back to petrol.

Now you have re calibrated with 3mm nozzles we need to see your 'model' again. Might be helpful to clear gathered plotted points in the 'map' screen, drive on petrol and let it regather plotted points, drive on gas and let it gather plotted points - and repost a pic of this screen too.

Simon

Iv'e set the live and earth connection from the battery terminals and the switch live from the petrol injectors to the LPG Ecu.

My mum and dad whent away for a couple of days, they've just come back and the eml was on and the code i got was "too rich bank 1", I have since corrected it and whent on a drive and i did'nt like the way the gas points where on the map as it was coughing and splutter and popping in the inlet on the odd ocasion, so i have gone into the map adjustments and made it so the gas points line up with the petrol map points. The car seems to respond better and doesn't cut off as much as it does
I have set the minium cut off pressure at 0.4 bar as this is the lowest i can set in the software.

I will get you some screen shots and post them up tomorrow night.

Cheers Luke

[Gilbertd]

it was coughing and splutter and popping in the inlet on the odd ocasion,

Popping and spitting back through the inlet is a sign of a weak mixture, not rich.

[luke123]

it was coughing and splutter and popping in the inlet on the odd ocasion,

Popping and spitting back through the inlet is a sign of a weak mixture, not rich.

I should of explained better that was happening without fine adjustments on the map

Cheers

luke

[luke123]

Here are some screen shots where im currently at with it, Iv'e also got a shot of the cut off problem showing what happens in the software.









Regards

Luke

[LPGC]

V30 injectors will be borderline working properly at 3.5ms pulse duration, they don't open quick enough.. This is with the warm(ish) gas temps shown on screen too - when the system compensates for colder (denser) gas (i.e. shortly after switching to gas in a morning), those pulse durations will decrease further still. On engine over-run (when all fueling isn't cut off) the figure will also be below 3.5ms... This assumes your Pinj's stay the same at idle when you switch between petrol and gas, I.e. assumes you have calibrated the mutiplier and corrections properly around the idle point - have you checked?

Noticed how all your numbers in the rpm correction screen are very similar? You could have set them all to zero and and made the same adjustment in the model (lifting the model line will increase the multiplier). Joining the dots should put you in the right direction following a recal... So, clear all the numbers in the corrections screen before you go on to the bit below...

As said, might not have been the best idea to drill the nozzles to 3mm but you should get away with it with the V30s and your engine. Now to get the injectors into their working range you will need to decrease reducer pressure - I would set reducer pressure to 1.1 bar, then when you do a full recal (not forgetting to set working pressure at 1.1 bar in the software too) you should see them doing around 4ms at idle and this will mean they're further into their working range. Then, join the dots, or even better drive with a steady foot at each ms point, switch between gas and petrol, note the difference and manually adjust the model at that ms point to keep Pinj the same.

Simon