Time to do some tuning

Message
Author
rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#21 Post by rodstares » Tue Jun 02, 2020 2:48 am

Frustrating. I hooked up the vacuum lines to that unused port and am getting 0.3 bar at idle, fitted my new temp sender that arrived from Tinley, fired her up all good. Damn thing will not calibrate! I've rebooted it multiple times, disconnected and reconnected power to the ECU, stopped and started the car. It says, "stabilize on idle and wait" seemingly forever. I can switch it to gas and it runs, switch back to petrol, no problem but no calibration.

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#22 Post by rodstares » Tue Jun 02, 2020 6:20 am

I got it to calibrate in the end. Seems like my idle speed is too high, over 1000 rpm in park so I turned on the headlights and heater fan which dropped the idle to about 980 rpm and away it went. Standby for a report.

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#23 Post by rodstares » Tue Jun 02, 2020 8:08 am

Alrighty, todays adventures... I got the vac hoses hooked up in the right position, the MAP sensor now reads 0.3 bar at idle, so far so good. I have the reducer vac line teed off the same spot. Got my reducer temp sensor from Tinley fitting and working. Had all sorts of trouble with autocal, finally figured out the idle on the Landcruiser is too high and it won't go into calibration until the revs are down under 1000 rpm which means putting the auto in drive or running the heater fan and headlights which reduces the rpm down to about 980 and then the calibration will kick off. So I eventually got calibrated at 1.3 bar gas pressure. It's driving fine but still switches back to petrol under load.
Using your method of holding my foot on the brake and revving it in drive at 2000 rpm gave me Pinj 12ms and Ginj 30ms. I have no lambda. That works out to 150% increase meaning my gas pressure needs to increase from 1.3 bar to 3.25 bar, that's too much, the reducer is only rated to 3 bar, can't be right. Something is pretty screwed up here and it might be my maths... I can increase the gas pressure with more shims under the spring but probably only to about 2.5 bar or so, any higher and I risk the PRV on the reducer blowing off and releasing gas under the bonnet.
The nozzle size is 3mm, I stuck a drill bit in them to clean out some fluff and dirt when I installed them.
So I guess, increase my reducer pressure, calibrate again and if still no good, petrol enrich or upgrade the reducer. I will go petrol enrichment first as I don't want to spend the money on a new reducer just now. Question is, can I drive it like it is without damaging anything?
Gas Display.PNG
Gas Display.PNG (114.2 KiB) Viewed 309 times
Gas Map.PNG
Gas Map.PNG (135.84 KiB) Viewed 309 times
Regards,
Rod

LPGC
Installer
Posts: 3865
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Time to do some tuning

#24 Post by LPGC » Tue Jun 02, 2020 2:04 pm

As it stands your multiplier figures (numbers in the boxes in the map table) will be very high because with the nozzles you have fitted in the injectors and running at 1.3 bar the injectors don't flow enough gas... In your 'display' screenshot above the 'injector nozzle size' bar clearly points to injector nozzle size being too small. Good installers don't always take much notice of the 'injector nozzle size' bar, it can depend on spec of the engine and spec of the injectors fitted too, but in this case we can safely assume that the bar showing injector nozzles are too small is correct!
[ Edit - Multiplier figures are indeed very high, in fact they're maxed out at 255... Which means the only way you could increase the amount of gas the system suppliers for any given petrol injector pulse length would be to increase the reference pressure in software from 1.3 to a higher figure]
This isn't a great situation regards correct fuelling even if you managed to get the figures in the map correct because it will mean other factors such as temperature correction don't work quite like how they should. It will also mean you won't be able to drive on gas with a combination of much above light/medium throttle at much above low rpms (because there won't be enough time between engine induction cycles for the injectors to open, flow gas and close again in time for the start of the next induction cycle. If your lambda voltage flicks properly while driving (various driving conditions) it won't hurt the vehicle to drive it like that but it will still switch back to petrol when you put your foot down a bit.

You would never want to increase pressure to much more than 1.7 bar, most types of injector don't work properly at more than 1.7 bar. The highest pressure systems work up to around 2.3 bar at most but they use a fairly unique type of injector which will work at that pressure, still those special injectors work better at around 2.1 bar. Best to assume max pressure your injectors will like is 1.7 bar. If you increased pressure to 3 bar then if the weather ever got to around freezing temperature (I know probably unlikely where you are) there'd be a strong chance of gas remaining a liquid as it entered the vapour lines, even if injectors could work properly at 3 bar (none can) they still wouldn't meter gas properly flowing liquid gas, they're designed to flow vapour gas. Much above around 1.9 bar would start to cause problems for your reducer (even if it could be adjusted to such high pressure).

When I refer to nozzles I mean the nozzles directly attached to the injectors as apposed to the nozzles that fit into the manifold. The manifold nozzles should have at least 3mm internal diameter, probably closer to 3.5mm or 4mm, they will (or at least should) always be bigger than the nozzles attached to injectors. Did you remove a nozzle from an injector and try pushing a drill bit through it? 3Mm nozzles on just about any make/model injectors should flow much more gas than the figures in your screenshots suggest! However, some injectors have internal restriction of less than 3mm (some such as certain batches of V30 injectors have 2.5mm internals), if injector internals are narrower than the nozzle the internals rather than the nozzle becomes the restriction. Just to confirm, what type of injectors do you have? Could you post a pic? Seems to me either you haven't measured the injector nozzle size properly (some types open up to 3mm at the end but the narrower restriction is further inside the nozzle toward the injector) or your injectors are of a type with a narrower than 3mm internal restriction. Is this a 4L straight 6 engine (cylinders 667cc each)? If so I'd expect your nozzles are closer to 2mm than 3mm. The standard shipping nozzle size on many types of injector is 1.9mm, you can buy bigger nozzles for some injectors but even where bigger nozzles are available off a shelf it's still fairly common practice for installers to buy standard size nozzles (that injectors are shipped with) and drill them themselves.

This may be obvious but worth saying that to drill injector nozzles you need to remove the nozzles from injectors first or the injectors will be damaged. But before you try to remove nozzles let's see a pic of injectors... Some types of injector are designed to have none removable nozzles, the nozzles are fitted at the factory and the extent to which they're tightened into injectors also affects the flow, it's possible to remove nozzles on this type but then it would be very difficult to calibrate injectors to match each other in terms of flow rate and response. Luckily there aren't many of this type of injector still around.

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

Gilbertd
Advanced Member
Posts: 8372
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Time to do some tuning

#25 Post by Gilbertd » Tue Jun 02, 2020 6:18 pm

I don't know much about injections systems other than the theory of how they work and something seems really wrong here. 3mm nozzles and 1.3bar gas pressure should be too much for your engine I would have thought. I once found a chart showing nozzle sizes, pressure and engine size but can't find it now.
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.

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#26 Post by rodstares » Tue Jun 02, 2020 10:13 pm

Good morning fellas. You make a good point regarding injectors. When I first had it running that size bar was right in the middle then it moved further to the too small end and even further after yesterday's recal. Might that indicate them be coming progressively blocked? Either way, sounds like I need to pull and inspect the injector nozzles. It was the injector nozzles I stuck a drill bit in, not the manifold nozzles but I never removed them from the injector so they could indeed have a restriction in the upper end. Today's job, I'll pull them and check them all. I believe they are valtek 34 or equivalent, I'll get some photos

LPGC
Installer
Posts: 3865
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Time to do some tuning

#27 Post by LPGC » Tue Jun 02, 2020 10:52 pm

Morning... ;-)

No, the bar reading moved further towards 'too small' because initially the pressure reading was 1.3 bar but the injectors were actually seeing 1.9 bar, now they're actually seeing 1.3 bar the injector flow rate relative to petrol injector flow rate seems even smaller. We had to connect the vacuum pipe to the correct point or you would never be able to calibrate it properly (or at least it would be very difficult to get consistent correct fuelling).

Before we moved the vacuum pipe there was atmospheric pressure on that pipe, so reducer output pressure was 1.3 bar above atmospheric pressure (not manifold pressure)... You still got a 1.3 bar reading because the map sensor was also reading atmospheric pressure. If we had only connected the map sensor to manifold vacuum and left the reducer vacuum port still at atmospheric pressure the pressure reading would have been 1.9 bar when manifold pressure is 0.3 bar. But we (correctly) connected both the map sensor and reducer vacuum port to manifold vacuum, so now the reading still shows 1.3 bar but this is now 1.3 bar above manifold pressure rather than 1.3 bar above atmospheric pressure. On this system (and most other systems) the gas pressure reading shows gas pressure minus manifold pressure.

But yes you do need to inspect injector nozzles, I expect they are too small.

Simon
Last edited by LPGC on Tue Jun 02, 2020 10:59 pm, edited 1 time in total.
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

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#28 Post by rodstares » Tue Jun 02, 2020 10:54 pm

The injectors
Inj2.jpg
Inj2.jpg (507.12 KiB) Viewed 290 times
inj.jpg
inj.jpg (402.97 KiB) Viewed 290 times
Inj 3.jpg
Inj 3.jpg (573.65 KiB) Viewed 290 times

LPGC
Installer
Posts: 3865
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Time to do some tuning

#29 Post by LPGC » Tue Jun 02, 2020 11:13 pm

They look like V34's but metal bodied, they have the Emer logo on the coils but iirc unlike most 34's these have the screw thread on the side.

Can you remove them from the car and unscrew just the lowest part of one of the nozzles, take more zoomed out pics from a few angles and measure the hole through the nozzle you removed (drill bit method)?

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

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#30 Post by rodstares » Tue Jun 02, 2020 11:26 pm

That's today's project. I have some actual work to take care of first this morning before I can start, which is a bit of a pain but sometimes I can't avoid it. Update to follow.
Cheers

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#31 Post by rodstares » Wed Jun 03, 2020 3:23 am

Alright, here are some pics of the nozzles, quite large the rivet I've got stuck in there is a slightly loose fit and it's 3.12mm so I would think the hole is around about 3.2 to 3.25, pretty large from what I am reading above. They injectors themselves look a bit dirty down in the bottom there where the nozzle seats. The hole down in the injector there where the nozzle seats is about 2.97mm and thats the minimum restriction, the nozzle is unrestricted 3.2mm or so.
inj4.jpg
inj4.jpg (498.89 KiB) Viewed 284 times
Inj7.jpg
Inj7.jpg (709.88 KiB) Viewed 284 times
inj6.jpg
inj6.jpg (367.95 KiB) Viewed 284 times
Last edited by rodstares on Wed Jun 03, 2020 4:56 am, edited 1 time in total.

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#32 Post by rodstares » Wed Jun 03, 2020 3:25 am

Inj5.jpg
Inj5.jpg (594.21 KiB) Viewed 284 times

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#33 Post by rodstares » Wed Jun 03, 2020 3:46 am

Looking at some charts I should have a around about 2.8mm nozzles for 30 kw/cylinder at 1.2 bar gas pressure, I am way over size on the nozzles

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#34 Post by rodstares » Wed Jun 03, 2020 5:02 am

Quite a bit of filth in the inlet to the injector bank too. Could the problem be caused by partially blocked injectors and oversized nozzels?
Inj inlet.jpg
Inj inlet.jpg (422.4 KiB) Viewed 283 times

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#35 Post by rodstares » Wed Jun 03, 2020 8:26 am

Well, I cleaned up the injectors by blowing some petrol through them, very nice. Recalibrated and took it for a run with the laptop connected. The software is a bit shit and it crashed twice, generally right on acceleration as the system switched itself back to petrol. The second time it did it the screen froze on the below. Interesting in that it shows the gas pressure has dropped to 0.8 bar which may indicate that the supply hose from the tank is too small (6mm) or the reducer is inadequate for the amount of flow required. I am starting to give those theories more weight. Anyway I have another 8mm gas hose here that may fit, I'll swap it out when I get the chance and see what that does. I looked at a brand new AMR Australian made reducer today, good looking unit, $320. Amazing, I didn't think we made anything here anymore. Anyway before I go down that route I'll try the hose.
Cheers gents.
Map test drive.PNG
Map test drive.PNG (134.17 KiB) Viewed 279 times

Brian_H
Intermediate Member
Posts: 1716
Joined: Mon Jan 23, 2012 12:21 pm
Location: Milton Keynes

Re: Time to do some tuning

#36 Post by Brian_H » Wed Jun 03, 2020 9:03 am

Given the amount of crud in the injectors, have you checked the filter in the solenoid housing to the reducer? If thats blocked up in a similar fashion that will cause problems with getting enough gas through.

LPGC
Installer
Posts: 3865
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Time to do some tuning

#37 Post by LPGC » Wed Jun 03, 2020 10:28 am

Usually if the feed to the reducer doesn't flow enough reducer pressure will fall off drastically above a certain engine load, or if the reducer itself won't flow enough pressure falls away a bit more gradually above a certain engine load... A feed issue will at some point lead to the reducer drawing vapour (cold vapour due to evaporation in the pipe or at the inlet filter), a reducer being fed with liquid one moment suddenly sees a sharp decrease in mass of gas at the input when the input becomes vapour... If a litre of liquid gas turns into about 270 litres of vapour at atmospheric pressure (1 bar) it will turn into 117 litres of vapour at 2.3 bar (1.3 bar above atmosphere when reducer manifold reference pressure is at 1 bar full throttle), a big enough reducer fed with a litre of liquid gas could turn that litre of liquid gas into the 117 litres of gas vapour at 1.3 bar above atmospheric pressure in a short time, no problem. The same reducer fed with gas vapour at (say) 3 bar (only 3 bar due to cold gas in the feed pipe due to evaporation in feed pipe) could convert a litre of gas vapour at 3 bar to only around 2.3 litres of gas vapour at 1.3 bar and since liquid gas at these pressures is 117 times denser in liquid form than vapour form it would take around 117 times longer to do it with vapour than with liquid (less some for difference in volume flow rate through reducer inlet metering jet between liquid gas and vapour gas, vapour will flow more volume than liquid but nowhere near enough to the extent that it could make up for the difference in density).

Presuming the injectors are flowing what we'd expect with 3mm nozzles and looking at rpm * ginj in the above pic, if this is a 4+ litre engine (what size is it again?) the reducer will be flowing as much gas as a 2L engine at full load (say 16ms ginj) at 4000rpm, I would expect the reducer to perform a bit better than that.... but we haven't yet talked about whether the mixture is correct... A very rich mixture could account for the below expectations reducer performance and account for the high figures in the map.

Have you checked mixture / fuel trims / lambda voltage? Autocal relies on the engine running closed loop mixture control, if for any reason the engine is running open loop autocal might enter very high or very low numbers in the map. In an earlier thread I said idle map pressure should be around 0.3 bar but I was later surprised to see that it was exactly 0.3 bar, on this engine I'd more likely expect 0.3odd bar - a rich mixture gives slightly more power for any given airflow than a stochiometric correct or lean mixture, so on an engine with electronic idle speed control if the engine idles with a rich mixture (at least up to a certain extent) manifold pressure at idle will be lower than it would be if mixture were correct.
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

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#38 Post by rodstares » Wed Jun 03, 2020 10:54 am

Brian_H wrote:
Wed Jun 03, 2020 9:03 am
Given the amount of crud in the injectors, have you checked the filter in the solenoid housing to the reducer? If thats blocked up in a similar fashion that will cause problems with getting enough gas through.
Yes mate, I put brand new elements in it and the gas filter

rodstares
New member
Posts: 114
Joined: Mon Nov 11, 2019 6:16 am

Re: Time to do some tuning

#39 Post by rodstares » Wed Jun 03, 2020 11:01 am

Thanks Simon, blown my mind again.... The engine is 4.5 litre straight six. Mixture/fuel trims/lambda is a mystery to me. This engine has no lambda, which is an oddity of the 1998 model 105 Landcruiser Aussie variant. The model prior had it and the model that started from 2000 onwards, but not this one.

LPGC
Installer
Posts: 3865
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Time to do some tuning

#40 Post by LPGC » Wed Jun 03, 2020 12:46 pm

rodstares wrote:
Wed Jun 03, 2020 11:01 am
This engine has no lambda, which is an oddity of the 1998 model 105 Landcruiser Aussie variant. The model prior had it and the model that started from 2000 onwards, but not this one.
Aha! In that case autocal cannot work properly and I expect your map (or at least most of it) is rich.

Autocal works by noting pinj (petrol injector pulse duration) when idling (normal or fast idle) and comparing to pinj when you switch to LPG. On an engine with an active closed loop fuel system if lambda reads rich the petrol ECU will decrease pinj to bring mixture correct and vice/versa, so if engine load is kept constant and fuel switched between petrol and LPG, any difference in pinj will be caused by the closed loop system correcting the mixture... For example if autocal is running with the engine idling with 3ms pinj on petrol then it switches to LPG and pinj falls to 2.7, autocal will assume that gas calibration is 10% rich so decrease figures in the map by 10%. But without active closed loop feedback (open loop - which your vehicle must be if it doesn't have a lambda probe) pinj will remain the same regardless of mixture so autocal will be working on incorrect pinj figures, which often leads autocal to set calibration either extremely rich or extremely lean.

LPG calibration on an engine with an open loop fuel system is more involved and more difficult to set correctly than it is on an engine with a working/active closed loop system but it can be done. Ideally you'd shove a wide band oxygen sensor into the exhaust and calibrate it while driving and monitoring mixture readings, next best would be to shove a narrow band oxygen sensor in the exhaust (again just at the tail pipe, or even better you could weld a lambda boss and fit a lambda sensor in the exhaust nearer the engine)..

But it can be done without any of these gizmos - By knowing the rough 'shape' of the map (the way the numbers in the map should start with comparatively medium figures, peak at around 4ms and the gradually decrease going further down the map, to suit the gas system injectors / nozzles / pressure versus engine cylinder size / cam profile / petrol injector flow rate and petrol pressure) where 'comparatively' means numbers relative to each other rather than absolute numbers, then (assuming the petrol system provides correct mixture) by manually comparing manifold pressure at idle on petrol to manifold pressure at idle on LPG, then change all numbers in the map (that we previously set correct shape for) by the same percentage so manifold pressure at idle remains the same on LPG as it is on petrol. Now assuming we got the shape of the map (transition of numbers going down the table from 1.5ms to 18ms) correct (remember this isn't about absolute numbers but is about how numbers compare to each other, it's about the percentage difference of numbers) and assuming petrol fuelling at idle is correct and assuming manifold pressure at idle on LPG as the same as manifold pressure at idle on petrol, our map should be correct. The problem with this method is that there are a lot of 'iffs' there and we haven't even touched on some of the other variables yet such as gas system temperature correction.

When you're comparing manifold pressure readings, remember that a little lean will usually lead to a small increase in map pressure, a little rich will usually lead to a small decrease in manifold pressure but a lot rich will also lead to only a small decrease in manifold pressure but a lot lot (very) rich will start to lean towards an increase in manifold pressure again.... So the best way to do it is slowly decrease all numbers in the map by the same percentage until manifold pressure on LPG equals manifold pressure on petrol, decrease calibration a bit more until manifold pressure rises to a bit more than it is on petrol (to confirm the correct point), then increase calibration figures back to the point manifold pressure on LPG equalled manifold pressure on petrol.

The best way for you to do it is probably a bit different to the 'setting the shape first' method I described above. It would probably be better for you to first warm the engine (fully) and let it idle on petrol, note idle manifold pressure, enter the figure 250 in every box in the table and switch to LPG, note the manifold pressure again, gradually decrease all numbers until you see manifold pressure start to rise. At what point (number in the table) did manifold pressure start to rise? All this is assuming your engine has electronic idle speed control! (??) If the engine doesn't have electronic idle speed control you'd set figures in the table according to rpm rather than manifold pressure.

You can easily select and change all numbers in the map at the same time, just click the box in the top left corner (1.5ms 500 rpm), hold down shift and click the box in the bottom right corner (18ms 6000rpm) and press enter. A box comes up in which you can set all figures you selected to an absolute value, linear will add any figure you enter to all the boxes selected (and you can add a negative value to decrease figures), percentage is like linear but adds/subtracts a percentage to all boxes selected. To start with you'll be using absolute to set all boxes to 250, the linear to add minus ten (-10) to every box (so the first time you add -10 with all boxes selected all the 250 figures will change to 240... probably sucking eggs but in case this adds to clarity).

When you've replied about what numbers in the map you got down to before manifold pressure started to rise (to above that on petrol) I'll reply with figures I reckon should be in the rest of the map, or you could even email me your map so I can make the changes for you and email it back to you.

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

Post Reply