Timing advance processor

[Pinger]

I watched that video on lean of peak in aircraft. Found it very interesting. Obviously cruising in an aircraft is a different situation than car on the a road with varying roads, hills etc. Going with what Pinger mentions about the Chevrolet forum guy is pursuing, I know on the TAP that I've got you can set a 3d timing map when you're hooked up to the MAP sensor also, so that would allow setting advance at all rev and load ranges. I like this idea, but honestly don't have the energy at the moment to put the energy into more basic things.

It is a massive undertaking if starting from scratch. The guys on the Chevrolet forum have a basic map to start with and are tweaking it in specific load/speed ranges. They are also doing this in conjunction with other (mechanical) changes eg, head work, different cam, etc - all of which makes re-tuning the ECU a necessity anyway.

[disco stu]

Cheers pinger. I don't even want to think about the work involved in that sort of thing. I used to get excited by those sorts of things, but energy is very lacking these days-I can't even remember what excited feels like!

That explanation of detonation is helpful and makes perfect sense, cheers.

On the mixture thing, where my thinking is sitting is that feedback from oxygen sensor could be affecting fuel trims (that is it's job in the end) changing the mixture......if things are different leaving the exhaust while using the TAP. Wish I was obd2 compatible so I could see what's going on in the ecu in regards to fuel trims etc, would take a lot of the guess work out if it

[Pinger]

On the mixture thing, where my thinking is sitting is that feedback from oxygen sensor could be affecting fuel trims (that is it's job in the end) changing the mixture......if things are different leaving the exhaust while using the TAP.

If the advanced timing is beneficial then the emerging exhaust gas should be cooler. I don't think lambda sensors are so very temperature sensitive re their outputs with temp but they will switch in and out with wider fluctuations in temp. Unheated sensors will switch in and out in in stop/go driving.
I only found out a week or so back that my heated sensors were being deprived their heating currents. Re-instated, the AFR gauge I have on the dash shows sensor output is a a lot more fluid.

[disco stu]

Well, just filled up to check consumption and it was at 16.4l/100km and wasn't flogging the car at all, so it's up 2l/100 from 2 fills ago at 0 extra advance. That was running 7deg extra advance. I've pulled it back to 3 and will see how this tank goes.

I'm guessing at 7 that it was before the optimum point and making the engine worka little harder

[LPGC]

Well, just filled up to check consumption and it was at 16.4l/100km and wasn't flogging the car at all, so it's up 2l/100 from 2 fills ago at 0 extra advance.

Assuming the tank fills to the same level each fill and liquid gas density(temp) is the same on each fill..

[Pinger]

Well, just filled up to check consumption and it was at 16.4l/100km and wasn't flogging the car at all, so it's up 2l/100 from 2 fills ago at 0 extra advance.

Assuming the tank fills to the same level each fill and liquid gas density(temp) is the same on each fill..

Beat me to it.
Fill quantity is such a variable that brim-to-brim calculations are nigh on impossible. PITA as it makes monitoring changes difficult. It is as I was told on this forum some time ago - a case of averaging consumption over several tankfulls and making sense of that as best you can.

[disco stu]

Yeah, good point. But I didn't think differences in density and angle of the car would make a difference of more than a few litres. Happy to be corrected though. Over half a tank or more that would be a fairly small difference in consumption figures, assuming I'm correct

[Pinger]

If my understanding and ability to explain is up to the task.... it's the vapour within the tank that needs to be compressed by the incoming liquid fuel for there to be room for the liquid. When the vapour is warm it resists compression (due to the higher (when warm) vapour pressure) and when the supplying pump hits its pressure cut-off the volume of vapour is quite high to the detriment of volume for liquid.
Easier understood when considered from the point of view of a very cold day and the vapour volume being tiny.

Even if my explanation is way off beam - the effect is significant.

[disco stu]

But the fuel pump cut off is from the float attached to fill valve rather than pressure, so therefore temperature difference would only affect the density of the lpg rather than the level at which the pump stops filling (assuming angle of the car is the same etc)

This is assuming everything is the same between uk and Australia, which I'm pretty sure it is after looking at multivalves from lpgshop and seeing the fill cut off floats

[LPGC]

Fill up on a hot day when liquid density is lowest and the pump will stop at a certain point, don't use any of the gas out of that tank but refill on a colder day when density of the liquid has increased and you might get a bit more gas in because the depth in the tank will have gone down a bit. This aspect only concerns density of the liquid but there are loads of variables that can effect how much gas a tank will take on a fill.

The design and capacity of the tank will also make a difference. If it's a 10 litre tank 2 litres variance would be a 20% variance, 1000 litre tank and it's only a 0.2% variance, more chance of density etc effecting the fill by 0.2% than 20%. Tank design (width vs depth) can effect things because it's the float mechanism (depth) that shuts off incoming gas, the float cut off mechanism may be consistent within tolerance of (say) 2% depth at any given liquid density but liquid density changes with temperature and this may effect float draft. Could talk about how pressure (pump vs tank) might effect the point at which the shut-off mechanism stops the tank filling, and how cold gas going into a hot tank slowly is going to have more time to be warmed by the metal tank before the tank is full compared to cold gas going into a hot tank quickly. Or how if the tank is completely empty and at low pressure when the pump is first connected the evaporation of liquid gas in the empty tank will cool it in the early stages of the fill. The pump may be designed to work at a certain pressure depending on temperature (to make for more consistent fill speeds as flow rate is pump pressure less tank pressure), if the pump doesn't shut-off itself the limiting factor in how much gas the pump can put in the tank can be pump pressure - if the pump is only pushing 8bar of pressure and the tank valve was fitted (tank sealed) with air in the tank then air pressure rises above the rising liquid LPG in the tank and air pressure can be the limiting factor.. e.g. if we seal a tank then fill it 90% full of liquid the air inside the tank is going to be compressed into 1/10th of it's original volume so will have 9bar of pressure over 1 bar atmospheric pressure, a pump at only 8bar over atmospheric pressure could only fill the tank to 8bar pressure.

[Pinger]

But the fuel pump cut off is from the float attached to fill valve rather than pressure, so therefore temperature difference would only affect the density of the lpg rather than the level at which the pump stops filling (assuming angle of the car is the same etc)

Good point.
Unless, that is the final cut-off and fuel supply can cease due to pressure before the fill valve gets to its closing point. I suspect though, I'm just digging myself deeper into a hole of my own making!

We need someone who better understands this. Thing is though, as a liquid, it won't be that compressible. I'm struggling to believe it is compressible enough for its density to alter significantly. Back to digging myself deeper into that hole. Have a look in your garden - I may pop up there.

[disco stu]

Cheers guys. I guess my point is that these variations aren't going to indicate consumption that is ~30% difference. I'm not measuring things down to the ml or anything, just trying to get ballpark figures. When I was around 14.5l/100 and then I get something like 17.5 then that's not going to be down to the change in temp when filling.

I will admit that this time when I filled I had only done 180km on the tank. If consumption was as before it would take 26l, but I put in 29.7 or something. So some/a lot of that difference could be from temperature differences

[Pinger]

What might be help in ascertaining if your TAP is helping you is acceleration runs over a speed/rpm range at different throttle openings.
You'd need an accurate TPS signal (a multi meter wired to its output would do) for the different throttle openings. Then, on the same stretch of road on the same day do back to back runs from say 1500 to 6000rpm in an intermediate gear with different advance settings and measure the time taken with a stopwatch (preferably with a split at say 4000rpm). You'd only need to run at 1/4, 1/2, 3/4 and WOT to get a picture of if advanced timing is working. Splitting at 4000rpm lets you see gains/losses at low/high rpm (which may otherwise cancel out).

[disco stu]

Being the nerd that I am I had to check the thermal expansion data on liquid lpg. It's listed as 0.003/k, or thereabouts depending on temp. Say a 10° difference (0.03 factor) with my 68 usable litre tank gives 1.95l.

Pinger: the software actually has a feature where you do basically that, but just put foot flat to the floor once it coasts down to say 1000rpm to measure time taken to get up to certain rpm. I wasn't seeing a great deal of difference between the graphs of different advance. The seller replied that I should have the graphs meeting differently to show difference. I was using 1000rpm, he said 3000 was better. I'll try and grab the image of that to put up.

Back and forth I was going over the same little bit of road revving the car out, was starting to worry the residents would call the cops! Need a quiet bit of road in the middle of nowhere

[Pinger]

Being the nerd that I am I had to check the thermal expansion data on liquid lpg. It's listed as 0.003/k, or thereabouts depending on temp. Say a 10° difference (0.03 factor) with my 68 usable litre tank gives 1.95l.

I just found a document online that has that kind of data - buried somewhere in its 256 pages....
Does 1.95l account for the different consumption?

Pinger: the software actually has a feature where you do basically that, but just put foot flat to the floor once it coasts down to say 1000rpm to measure time taken to get up to certain rpm. I wasn't seeing a great deal of difference between the graphs of different advance. The seller replied that I should have the graphs meeting differently to show difference. I was using 1000rpm, he said 3000 was better. I'll try and grab the image of that to put up.

I thought I saw you post that earlier - but WOT only?
You could be overlooking part throttle gains (where most driving is done) and where gains are more probable IMO.

Back and forth I was going over the same little bit of road revving the car out, was starting to worry the residents would call the cops! Need a quiet bit of road in the middle of nowhere

Less obtrusive at part throttle.

[disco stu]

1.95l isn't as much as the total difference I saw, but considering temps when I filled each time I actually would have expected to put less in this time around.

That's a good point on part throttle vs wide open. You think some of the gains might only be seen at part throttle in same rev range?

I could even set up some sort of a temporary stop underneath throttle for the tests so I don't vary from that position. Bit scared to rev the thing out to 6000 though!

I still like the idea of monitoring exhaust temps also. I'm pondering this one

[Pinger]

That's a good point on part throttle vs wide open. You think some of the gains might only be seen at part throttle in same rev range?

I could even set up some sort of a temporary stop underneath throttle for the tests so I don't vary from that position. Bit scared to rev the thing out to 6000 though!

Very possibly as at WOT timing is retarded. It's in the part load situations where gains can be found. And, as that's where most of your driving occurs, any gains there are better exploited.
If you are going to 'stop' the throttle - try and do it at the engine side as accuracy here is critical. Does your LPG diagnostic/set-up software display TPS output in real time? If it does, use that. In 1 Volt increment starting at 2.0V for 1/4 throttle. 5.0V is WOT so 3.0 and 4.0V will show 1/2 and 3/4 of WOT. Repeatability is everything here. And you have the choice of rpm range to test over. Higher than the red line is pointless, low rpm where you'd downshift to avoid can get less attention.

I'd kind of expect an engine without a knock sensor to be quite conservatively timed - and LPG has better octane number than the petrol the engine was configured for. It's mapped settings will probably take an across the board advancement.

I still like the idea of monitoring exhaust temps also. I'm pondering this one

Tricky as back to back comparisons are required. Measuring acceleration by contrast is a measure of engine output - merely using the vehicle as the inertial part of an inertia dyno.

[Budgetbond]

I appreciate the intricacies of your mapping dilemma dstu.

How mappable is your TAP, I assume you can vary the advance across the full rpm range if you further wanted too.

I've been looking into the revs I normally use so I sort of know at which points it's good to get efficient calibration.
For me 2-3000 rpm equates to 50-70mph, so cruising at 56mph mite see some gains with a tap, at towards 70 and above you'd probably be reaching the break even point on my engine maybe.

However most of my driving is below this, where it prolly counts more.

What are your usual driving rpm s most important for u to optimise?

Can appreciate the difficulty s in getting feedback/stats to base your adjustments on, I suppose if you avoid filling up on unusually warm or cold days from the same pump may help a bit.

Like u say it does sound like 7 degrees led to increased consumption probably.

Obviously you can just use trial and error feel on consecutive tankfuls used in normal driving to try and hone it down, perhaps another better way of monitoring fuel consumption could be to set-up your fuel gauge light voltages exactly linearly and time or check miles between lights extinguishing at an approx constant cruising speed on long runs. (Or checking fuel gauge voltages somehow)

I'm trying to think as a total non-expert how else you mite gauge efficient advance running.

Perhaps monitoring something whilst switching from between the fuels?

Perhaps if you could get some form of magnetic temperature gauge you could put on the exhaust manifold?

I would imagine from 1-3000 rpms are the only parts that are worth trying to configure accurately, I don't suppose like me you go over 4000 rpms much anyway.

Soz just throwing outta couple of extra random ideas/thoughts there

[disco stu]

Sorry for delay replying here. Been busy and forgot to get on.

So basically, I've run it at 0deg advance for a few tanks to make a good comparison.

A while back after oxygen sensor was replaced, I was getting 12.5l/100km on petrol, and got one tank of LPG at 14.5l/100km. I was really happy with that and was hoping to stay around those figures. Since trying the TAP, figures have gone up, but even now at 0deg advance I'm still getting higher consumption than I was hoping. I feel the torque was a little higher when running some advance, and wouldn't mind a little more but fuel consumption is my main priority.

These are some of the figures I've gotten at various advances, all in L/100km:

9deg advance (I think) 17.75
7deg 16.4
3deg 16.5
0deg 15.8
0deg 16.1

Obviously none of these figures are even real close to the 14.5L/100km that I did get. Conditions for driving are fairly similar for all with majority long distance hwy driving, though I do the occasional trip into town and battle the traffic lights. I drive conservatively. I'll see what I get on the next fill, but it looks like 0 advance is giving the lowest fuel consumption. Still tempted to try 1 or 2deg in case its the sweet spot.

What sort of % increase in consumption would you expect between petrol and LPG? I had read that 15-20% was expected, and lines up with energy content vs volume difference. LPG at 14.5 is 16% higher which was around what I saw originally when I first got it going and O2 sensor was stuffed. At 16l/100km it is 28% higher, which is getting beyond just energy content differences.