Pipe sizes.

Message
Author
LPGC
Installer
Posts: 4511
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Pipe sizes.

#261 Post by LPGC » Wed Jan 12, 2022 11:32 pm

The ideal stepper/default position is different for every engine and with every mixer/reducer combination.

The general advice for the 175 based systems was to aim for a default position of around 80 (around 1/3rd opening) but that won't work well on all setups.

On a perfect and easy to set up system the stepper doesn't need to move much under any conditions (e.g. not for idle, or over-run, or under low / medium / high loads) but mixers don't create an entirely linear vacuum signal to the reducer for varying airflows and reducers don't have an entirely linear response to signal... And the ideal mixer will only cause a minimum drop in airflow (minimum drop in bhp) but that also means it's likely to produce a below linear signal at minimum airflow - Suppose we had a 1metre diameter pipe and blew air through it at a vast rate, if we stuck a narrower pipe with a shield (e.g. the end of the pipe was cut at an angle so the upstream end shielded the shorter unshielded end) through the side of the 1m pipe we'd get a vacuum signal that might be quite linear if we increased airflow further or decreased airflow a bit, but if we tried just blowing/exhaling using our breath through the end of the 1 metre diameter pipe we wouldn't see any signal on the side pipe because the airflow wouldn't be enough to create any signal at all, the breath wouldn't be enough to create any sort of signal because it would be completely off the bottom end of airflow at which the shield type mixer would generate signal... to the reducer there would be no difference between zero airflow and us blowing through the pipe. But too small a mixer will start to see the gas outlet in the mixer see manifold vacuum as engine load increases, so we might be able to get good signal response at idle and low engine loads but at higher loads the tendency will be to go way too rich almost regardless of reducer sensitivity or inline adjuster settings because under the manifold vacuum conditions the gas outlet will see much higher vacuum than the vacuum that any normal venturi or shield vacuum would ever reach . All this may seem intuitive enough, but it becomes less intuitive when you realise that 2 mixers with equal size venturi may flow the same at high engine loads but one mixer may remain more linear at lower engine loads than another.

There are guides etc for setting up mixer systems but they're often written by someone who's set up just a few mixer systems, may never have had to fit a slimline (less linear at low airflows) mixer because a full size mixer won't fit if the bonnet is going to close, may never have come across a situation where it is necessary for stepper position at one extreme of engine load to be quite different to the position it needs to be for the other extreme of engine loads.

As said, the simplest situation is if the stepper hardly has to move, where possible installers back in the day used to love to see that at an early stage during setup because it was an indicator that the install was going to be straightforward and tuning a doddle. But the engine could probably have made a few more horses foot down if the stepper did have to move a bit between idle and high load because if we have enough signal at very low loads to mean that the mixer's response is totally linear it usually also implies that we could have fitted a mixer that was slightly less restrictive to airflow and increased reducer sensitivity to signal a bit... So we could have fitted a bigger mixer, increased reducer sensitivity a bit, and then expect the mixer to be less linear at idle and the stepper position to need to open a bit for enough fuelling at idle but the better flow of the bigger mixer could mean less restriction to airflow and less restriction to peak bhp.

Then we return to the subject of default stepper position. We need to remember that for a given rpm and TPS the stepper motor will only move at a certain speed, the speed of movement of the stepper normally increases with increasing rpm reading but the current position of the stepper motor doesn't affect the speed at which the stepper motor moves. If we have a stepper motor position of 200 and the lambda reading is lean, then a single stepper motor position increase will move it to a position of 201, which in simple terms equates to a 0.5% increase in the stepper's position. If the position was 100 and we increased it to 101 there would be a 1% increase... in simple relative terms twice as much of an increase as the step increase from 100 to 101 in the same amount of time as it would otherwise increase from 200 to 201, so the system would be likely to correct for the lean mixture in half the time with default set to 100 as with default set to 200. The point here is that stepper position also affects the speed at which the system can adjust mixture and compensate for lambda readings but on the flip side too much speed of adjusting mixture can result in loss of fuelling accuracy under certain conditions at constant load.

There are many knock on effects of default position, the best all round compromise should be aimed for (though the last thing you want is for the engine to lean out and die at idle) but the best compromise may depend on the setup and engine. E.g. If we connected to TPS then when we move the throttle position the stepper jumps back to the default, for good driveability power and fuel economy we need the default to be close to the ideal for load conditions... but then if we have a mixer that is unrestrictive for load conditions so good for peak bhp the engine may return to a lean idle until the stepper moves up to correct mixture at idle if the engine doesn't die. In this condition we could lock the default at a higher position so moving the throttle under load would see the engine initially slightly rich (better for driveability and economy than lean) and then the engine wouldn't return to as lean an idle or die. Sometimes you'll get a setup on which things are the other way around, lower position for idle than for load even with the bypass closed... So first you'd check for the ram air effect leaning mixture while driving forward but then you might come to one of many conclusions including the possibility of the mixer being too small. You'll also get setups that go rich or lean on the over-run - going rich on over-run so the stepper closes can be OK but over-run usually means closed throttle and closed throttle will remain when the engine needs to idle again, so watch out that the stepper isn't in too low a position that causes the engine to stall on return to idle (maybe limit closing positions). Going lean on over-run (the opposite) could mean the engine has problems running due to too rich a mixture (but less likely). Another problem with going too lean on over-run is the likelihood of backfires... You can get the same problem using the 'cut-off' feature. Some installers used to set the position for the stepper during cut-off to zero thinking that would prevent the engine getting any gas during the operating condition, it doesn't mean the engine won't get any gas, the solenoids remain open, it does usually mean the engine will get a very lean mixture and will be pumping gas into the exhaust during the condition and may cause a big bang on a long enough over-run or return to power immediately after the condition,

For all of the intricate design there's nothing really special about a mixer, they create a restriction to airflow and cause a vacuum signal which 'sucks' gas out of the vaporiser.. but to read some of the old manufacturers blurbs you'd think they could defy the laws of physics. There is no way that a hole in a mixer 20mm narrower in diameter than a straight pipe inlet duct or throttle body could flow as much air as the inlet but to read some blurbs it seems they're trying to say that by designing the restriction in the mixer to be aerodynamic they can make for no airflow losses through the mixer. I've driven thousands of miles using a homemade mixer made from a baked beans can with a carefully sized hole in one end and a gas pipe stuck through it's side and it performed with no perceptible different to a manufacturer's mixer. You don't even need a venturi (hole) to make a mixer, you can just shove a pipe into an inlet duct with a shield on the airflow upstream side. The venturi type and the shield type both have pros and cons and in some respects the pros and cons of the respective types are opposite to each other - I reckon that with a decent vaporiser it would be possible to make for a setup on most installs where the stepper wouldn't have to move much (all round better linearity) and the overall setup would be less restrictive to either type standalone, by fitting both types i.e. an oversized mixer and undersized shield type with a shared connection to the vaporiser. But these days not something worth pursuing, the range of vehicles we can fit mixers on isn't getting bigger. I do like mixer systems though, in many ways they overcome the issues petrol fuel injection systems were designed to overcome and have the default advantage of providing every cylinder with equal mixture regardless of valve/cam gear wear which can see some cylinders get more air than others and with much less complexity to an injection system, but the necessary restriction to airflow is a minor negative. I always felt better when I had to crank reducer sensitivity up high and had fitted a full size (good signal at low loads for given venturi diameter) mixer... then I knew I couldn't have fitted any bigger less airflow restrictive mixer so the mixer I'd fitted would restrict bhp to the least possible extent.

I also like the gas carb setups like Impco, they really are like a carb but for gas, gas is sent to the carb unit under pressure, the gas carb itself has a diaphragm and the main gas metering apparatus not the vaporiser. In theory no better than a mixer but in practice they're better particularly during sudden load changes.
Last edited by LPGC on Thu Jan 13, 2022 1:08 am, edited 3 times 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

Hairyloon
New member
Posts: 147
Joined: Thu Mar 18, 2021 10:40 pm

Re: Pipe sizes.

#262 Post by Hairyloon » Wed Jan 12, 2022 11:45 pm

Am I going to regret it if I ask what you mean by a shield?

I think (unless it's blind hope) that the answer now is to drive it about a bit and see how it goes.
Probably I should set something up to give me the Lambda reading on the dashboard, but I think I ought not wait for me to get around to doing that...

Gilbertd
Advanced Member
Posts: 8773
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Pipe sizes.

#263 Post by Gilbertd » Wed Jan 12, 2022 11:51 pm

Laptop in the passenger footwell should show you what the system is doing. You don't need to stare at it all the time, not a good idea, but a glance whenever conditions change.
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: 4511
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Pipe sizes.

#264 Post by LPGC » Thu Jan 13, 2022 12:30 am

Don't go trying to change your setup just because I mentioned shield type setups...

But to explain, shield types are just a pipe stuck into the airflow with a shield (plate etc) at the side of the pipe from the direction of where the airflow is coming, the shield creates a lower pressure area in the areas of the outlet of the pipe, the lower pressure creates a similar vacuum signal to the venturi of a mixer.

As Gilbert said/implied, for your closed loop system to work properly you must have wired it to lambda and in the LPG software lambda signal voltage will be visible on the laptop screen. You wouldn't want to stare at the screen for long, especially not if you're doing say 160mph on a local bypass...
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

Hairyloon
New member
Posts: 147
Joined: Thu Mar 18, 2021 10:40 pm

Re: Pipe sizes.

#265 Post by Hairyloon » Thu Jan 13, 2022 1:23 am

Gilbertd wrote:
Wed Jan 12, 2022 11:51 pm
Laptop in the passenger footwell should show you what the system is doing. You don't need to stare at it all the time, not a good idea, but a glance whenever conditions change.
I was going to ask a friend to come along... though the preferred friend for the task has said no. Believde it or not, I do have another friend...
IMG_20210605_204052.jpg
IMG_20210605_204052.jpg (752.18 KiB) Viewed 195 times

More seriously though, I was planning to stick a microcontroller in the van to monitor air pollution and stuff. It ought not be difficult to have it give a readout on the dash as well.
LPGC wrote:
Thu Jan 13, 2022 12:30 am
Don't go trying to change your setup just because I mentioned shield type setups...
If she starts from cold tomorrow, I'm changing nothing until I've got a clear idea of what's going on... unless it runs like a bag of bolts.
You wouldn't want to stare at the screen for long, especially not if you're doing say 160mph on a local bypass...
If my van gets up to 160, then the lambda readings will be the last thing on my mind.

Hairyloon
New member
Posts: 147
Joined: Thu Mar 18, 2021 10:40 pm

Re: Pipe sizes.

#266 Post by Hairyloon » Thu Jan 13, 2022 1:28 am

LPGC wrote:
Wed Jan 12, 2022 11:32 pm
The general advice for the 175 based systems was to aim for a default position of around 80 (around 1/3rd opening) but that won't work well on all setups.
I'd understood that the ECU sets its own default based on what it has learned...

LPGC
Installer
Posts: 4511
Joined: Wed Sep 29, 2010 6:01 pm
Location: Yorkshire
Contact:

Re: Pipe sizes.

#267 Post by LPGC » Thu Jan 13, 2022 10:22 am

It does learn/set it's own default but you'd expect something else would need adjusting if the default position were near either extreme range of movement (near fully closed or fully open), if it were near an extreme it wouldn't be able to move much further before reaching the extreme. If you don't want it to be near either extreme it's kind of intuitive that you want it somewhere close to the middle.

I believe the old advice of ideal position to be at around 80 (which is around 1/3rd open) was to increase the speed at which it could react to adjust mixture and to err on the side of caution regards potential for lean mixtures.
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: 8773
Joined: Wed Apr 09, 2008 10:00 am
Location: Peterborough

Re: Pipe sizes.

#268 Post by Gilbertd » Thu Jan 13, 2022 11:09 am

Hairyloon wrote:
Thu Jan 13, 2022 1:28 am
I'd understood that the ECU sets its own default based on what it has learned...
It does, but where the default sits will be dictated by the vaporiser settings. If it needs to be open more than you want, slacken the vap bias off to allow more gas for any given amount of 'suck' from the mixer. That's where the juggling act comes in, you want it so the opening is the same, or as close as you can get it, at idle and out of idle.
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.

Hairyloon
New member
Posts: 147
Joined: Thu Mar 18, 2021 10:40 pm

Re: Pipe sizes.

#269 Post by Hairyloon » Thu Jan 13, 2022 11:15 am

I had it running fairly well yesterday, but then this morning wouldn't start from cold. Tried it on petrol and it's doing the same thing as before, where it's dying off when you open the throttle... maybe the problem is the ignition? Vacuum advance not advancing properly perhaps?
Something to have a look at anyway.

I need to double check the wiring again: see if I can find why it's not lighting up when we think it ought to.

Post Reply