Where can i learn how to map my own car.
Discussion
Hi Adrian
What sort of schedule would you think for testing to rule out variations etc on an engine dyno, specifically our SF901 water brake? We were thinking of allowing several minutes at each rpm to stabilise, then idle for a few minutes, then next 100 rpm increment etc etc, maybe three hours for a run? Any input would be most appreciated.
I suppose one of the problems doing such steady state testing is time scale. Three runs for statistical nicety, that is one day gone. Then if one wants to experiment......lots more weeks and months!
With regard to smoothing graphical output, I think some folks fail to realise even analogue gauges are damped to give smoother output so nearly all data is smoothed in some way and not raw warts and all stuff.
Peter
What sort of schedule would you think for testing to rule out variations etc on an engine dyno, specifically our SF901 water brake? We were thinking of allowing several minutes at each rpm to stabilise, then idle for a few minutes, then next 100 rpm increment etc etc, maybe three hours for a run? Any input would be most appreciated.
I suppose one of the problems doing such steady state testing is time scale. Three runs for statistical nicety, that is one day gone. Then if one wants to experiment......lots more weeks and months!
With regard to smoothing graphical output, I think some folks fail to realise even analogue gauges are damped to give smoother output so nearly all data is smoothed in some way and not raw warts and all stuff.
Peter
PeterBurgess said:
Hi Adrian
What sort of schedule would you think for testing to rule out variations etc on an engine dyno, specifically our SF901 water brake? We were thinking of allowing several minutes at each rpm to stabilise, then idle for a few minutes, then next 100 rpm increment etc etc, maybe three hours for a run? Any input would be most appreciated.
I suppose one of the problems doing such steady state testing is time scale. Three runs for statistical nicety, that is one day gone. Then if one wants to experiment......lots more weeks and months!
With regard to smoothing graphical output, I think some folks fail to realise even analogue gauges are damped to give smoother output so nearly all data is smoothed in some way and not raw warts and all stuff.
Peter
That's really more a max question - if the control system isn't oscillating madly, it should stabilise in a handful of seconds at worst.What sort of schedule would you think for testing to rule out variations etc on an engine dyno, specifically our SF901 water brake? We were thinking of allowing several minutes at each rpm to stabilise, then idle for a few minutes, then next 100 rpm increment etc etc, maybe three hours for a run? Any input would be most appreciated.
I suppose one of the problems doing such steady state testing is time scale. Three runs for statistical nicety, that is one day gone. Then if one wants to experiment......lots more weeks and months!
With regard to smoothing graphical output, I think some folks fail to realise even analogue gauges are damped to give smoother output so nearly all data is smoothed in some way and not raw warts and all stuff.
Peter
How long you want to hold the engine for boost/torque to stabilise after that, I can't really help you.
re damping :
Since a 4 cylinder engine goes though a torque reversal every cycle, obviously some smoothing is necessary. How much is a cause of many arguments
.I would rather see the lumps and bumps, but I don't tune cars for money, and am just as interested in control system artefacts as dips in the torque curve. Not to sound boastful, but on a dyno I know intimately, operated by me, I can generally tell which is which.
I also only play with NA engines, which have nowhere near as savage torque spikes.
I hate to sound cynical, but I am very glad I don't have to deal with some of our customers' customers. So many want curves like the ones in magazines or brochures. In my limited personal experience with cammy 4 cylinders, that nice smoothed torque curve is illusory, and hiding potentially useful information.
Thanks for the info Adrian, maybe Max will shed some more light on the engine dyno testing.
With regard to smoothing, we stick to level 4 out of 5 which also gives us a 'level' playing field. The spikes can and do skew max bhp and max torque upwards. I know of a few dyno places where the operator goes through the levels of smoothing to find the one with max bhp and torque and prints that out for the customer. As I said earlier though, we do go through the smoothing levels to see what is happening and on a few occasions have run the cars/bikes in lower/higher gears to see if anomalies are speed related or rpm related.
Peter
With regard to smoothing, we stick to level 4 out of 5 which also gives us a 'level' playing field. The spikes can and do skew max bhp and max torque upwards. I know of a few dyno places where the operator goes through the levels of smoothing to find the one with max bhp and torque and prints that out for the customer. As I said earlier though, we do go through the smoothing levels to see what is happening and on a few occasions have run the cars/bikes in lower/higher gears to see if anomalies are speed related or rpm related.
Peter
I had a guage to monitor back pressure next to my boost guage when I ran a big vgt Holset and would see roughly 2-1 ratio boost -back pressure with the vane in its smallest position coming down to a few psi apart at 1.8 bar with fully open vane at 2.0 a/r . Pity I couldn't get it to fit with the Lexus v8 as it would have been perfect for the high comp ratio
I had a guage to monitor back pressure next to my boost guage when I ran a big vgt Holset and would see roughly 2-1 ratio boost -back pressure with the vane in its smallest position coming down to a few psi apart at 1.8 bar with fully open vane at 2.0 a/r . Pity I couldn't get it to fit with the Lexus v8 as it would have been perfect for the high comp ratio
AW111 said:
Silent1 said:
Thanks chaps, that sounds like a good starting point, I think an aftermarket ecu might be a good idea as then as I move into other cars (of my own) I can take some knowledge with me.
One advantage of reprogramming a factory ECU is you get to keep all the corner mapping : below zero ambient, hot start, etc. and only change what you want to. A new ECU needs a base map from somewhere.The main thing I see is time : if you only have a budget for an hour of dyno time, you will probably ask the tuner for max power. If you want all the transients and part-throttle mapping optimised as well, you will be paying for a lot more time.
I still intend on hopefully paying someone to walk me through it to begin with, i don't intend on just diving in and f
king it up 
Hi Adrian
Have you run many single cylinder 4 strokes on the rollers? I attach 1 smoothing and 4 smoothing graphs for a Yamaha 660 big single. The graphs show with (blue) and without (green) the snorkel which fits under the fuel tank. It makes the engine run a little rich. I would have thought a single cylinder would have been affected to a great extent by pulses and tuned length. To be honest this does not seem to be the case and we reckon the power difference can be attributed to the reduction in air flow restriction reducing signal on main jet weakening fuel mix. We do recall a reduction in power below 3250 and an increase 3250-3800 then a small increase everywhere till 4800 when we see another hump in power to 6000. Maybe this is the 'pulse' tuning effects? Overall, even with the nobbly offroad tyres, cush drive, roller chain and extreme reversions of torque on a single cylinder, the graphs, even the minimum smoothing one are surprisingly even.
Peter
Have you run many single cylinder 4 strokes on the rollers? I attach 1 smoothing and 4 smoothing graphs for a Yamaha 660 big single. The graphs show with (blue) and without (green) the snorkel which fits under the fuel tank. It makes the engine run a little rich. I would have thought a single cylinder would have been affected to a great extent by pulses and tuned length. To be honest this does not seem to be the case and we reckon the power difference can be attributed to the reduction in air flow restriction reducing signal on main jet weakening fuel mix. We do recall a reduction in power below 3250 and an increase 3250-3800 then a small increase everywhere till 4800 when we see another hump in power to 6000. Maybe this is the 'pulse' tuning effects? Overall, even with the nobbly offroad tyres, cush drive, roller chain and extreme reversions of torque on a single cylinder, the graphs, even the minimum smoothing one are surprisingly even.
Peter
Edited by PeterBurgess on Sunday 1st November 09:07
Peter : a couple of thoughts :
If I didn't point this out, max would for sure
One question : with a change in intake tuning I'd expect more crossovers in the graphs- better in some places, worse in others; or an overall drop in performance. Did you log AFRs?
Although I'm happy to discuss this stuff, I have to reiterate that my experience is in the control system side, and although I work with some excellent tuners, I am a hobbyist at it personally. Others one here, including yourself, have a lot more tuning experience than me. On the other hand, I have been doing this on and off for over 20 years, seen a lot of graphs and the raw numbers behind them.
When I'm older and greyer, I will publish a rant on "How not to use your dyno", with topics including
"If the tyres are on fire, it's probably wheelspinning",
"If it only reads wrong in one particular car, it's probably not the dyno" and
"Just because your car does 200 mph, you don't have to dyno at that speed"
OT
My latest project is a new exhaust manifold for the FIAT twincam rally car. 1967 and not a transistor in sight!
As is always the case, theory meets practical limits : there's a steering box and sump guard to be negotiated, so there have to be compromises
- I don't like the effect the smoothing has on the second graph : it changes the shape of the torque curve considerably.
- Contrary to what you might think, the chain, tyres, clutch hub, suspension sag : they all help with running a single cylinder 4 stroke. We have built a few go-kart engine dynos, and everything vibrates loose, or fatigues and cracks
. They are a horrible thing, but then so are the 18,000 rpm 2 strokes, which also try to vibrate everyting yo destruction.
If I didn't point this out, max would for sure
One question : with a change in intake tuning I'd expect more crossovers in the graphs- better in some places, worse in others; or an overall drop in performance. Did you log AFRs?
Although I'm happy to discuss this stuff, I have to reiterate that my experience is in the control system side, and although I work with some excellent tuners, I am a hobbyist at it personally. Others one here, including yourself, have a lot more tuning experience than me. On the other hand, I have been doing this on and off for over 20 years, seen a lot of graphs and the raw numbers behind them.
When I'm older and greyer, I will publish a rant on "How not to use your dyno", with topics including
"If the tyres are on fire, it's probably wheelspinning",
"If it only reads wrong in one particular car, it's probably not the dyno" and
"Just because your car does 200 mph, you don't have to dyno at that speed"
OT
My latest project is a new exhaust manifold for the FIAT twincam rally car. 1967 and not a transistor in sight!
As is always the case, theory meets practical limits : there's a steering box and sump guard to be negotiated, so there have to be compromises
AW111 said:
Peter : a couple of thoughts :
If I didn't point this out, max would for sure
One question : with a change in intake tuning I'd expect more crossovers in the graphs- better in some places, worse in others; or an overall drop in performance. Did you log AFRs?
Although I'm happy to discuss this stuff, I have to reiterate that my experience is in the control system side, and although I work with some excellent tuners, I am a hobbyist at it personally. Others one here, including yourself, have a lot more tuning experience than me. On the other hand, I have been doing this on and off for over 20 years, seen a lot of graphs and the raw numbers behind them.
When I'm older and greyer, I will publish a rant on "How not to use your dyno", with topics including
"If the tyres are on fire, it's probably wheelspinning",
"If it only reads wrong in one particular car, it's probably not the dyno" and
"Just because your car does 200 mph, you don't have to dyno at that speed"
OT
My latest project is a new exhaust manifold for the FIAT twincam rally car. 1967 and not a transistor in sight!
As is always the case, theory meets practical limits : there's a steering box and sump guard to be negotiated, so there have to be compromises
Thanks for the earlier reply, it does make sense. Funny you know, the engine dyno graph (with the usual dip in it) I was going to post was the old FL twin cam too although not as old as that! Which one have you got?- I don't like the effect the smoothing has on the second graph : it changes the shape of the torque curve considerably.
- Contrary to what you might think, the chain, tyres, clutch hub, suspension sag : they all help with running a single cylinder 4 stroke. We have built a few go-kart engine dynos, and everything vibrates loose, or fatigues and cracks . They are a horrible thing, but then so are the 18,000 rpm 2 strokes, which also try to vibrate everyting yo destruction.
If I didn't point this out, max would for sure
One question : with a change in intake tuning I'd expect more crossovers in the graphs- better in some places, worse in others; or an overall drop in performance. Did you log AFRs?
Although I'm happy to discuss this stuff, I have to reiterate that my experience is in the control system side, and although I work with some excellent tuners, I am a hobbyist at it personally. Others one here, including yourself, have a lot more tuning experience than me. On the other hand, I have been doing this on and off for over 20 years, seen a lot of graphs and the raw numbers behind them.
When I'm older and greyer, I will publish a rant on "How not to use your dyno", with topics including
"If the tyres are on fire, it's probably wheelspinning",
"If it only reads wrong in one particular car, it's probably not the dyno" and
"Just because your car does 200 mph, you don't have to dyno at that speed"
OT
My latest project is a new exhaust manifold for the FIAT twincam rally car. 1967 and not a transistor in sight!
As is always the case, theory meets practical limits : there's a steering box and sump guard to be negotiated, so there have to be compromises
Evoluzione said:
Thanks for the earlier reply, it does make sense. Funny you know, the engine dyno graph (with the usual dip in it) I was going to post was the old FL twin cam too although not as old as that! Which one have you got?
The FIAT is not mine, but I have been navigating and helping with prep for years. It's a 1967 1500 mk3, but we have papers to prove a few ran the twincam in period. The engine is 1736 cc if I recall correctly, running twin downdraft webers.And no, I don't tune it : webers are best left to people with lots of experience (and jets, chokes, emulsion tubes, and all those doohickeys).
Sorry Adrian no logged afrs, we had broken the Powerdex connected to the dyno and were using the Innovate LM2 back up which is not connected to dyno. We bought two new Powerdexs and thus have one as a spare in case it gets broken again! Basically the snorkel made it run rich. This is backed up by my mate who uses a Dynojet bike dyno, we could never understand why the snorkel was fitted as the bike must have been jetted without snorkel! We did not alter fuelling, only testing OEM vs remove snorkel as the owner has read more bhp if remove snorkel. Speaking of which, the number of times we tune cars which are too rich cos they have been engine dynoed or rolling road dynoed without air filters for some reason, take off the filters and the mixtures are near enough right.
Peter
Peter
The op hasn't said who he used to use, it sounded like 1320 but I guess maybe not. They use BytetroniK, and they use the Full Access kit which allows you to do everything they do, and they'll run through it with you. Also BytetroniK have their own forum for international assistance. It uses the OE ecu for the R53 which is what the system was designed for. I have it, and it is great, but I haven't altered anything yet.
The system records live data, and is designed to work with Innovate systems, so it will record wide band data, essentially removing the need to use a dyno to record AFR. There is a dataviewer which can be purchased separately, but is good to highlight a specific point in the map to concentrate on. I fitted the wide band sensor, and have the gauge in the front cup holder which fits perfectly.
Just my2p, you'd be welcome to sit in the car at some point to gain an idea, it isn't a race car, just fast road suppose, but 250chp with standard exhaust cat back, standard I/C, and standard cylinder head currently, the car looks standard in every way.
The system records live data, and is designed to work with Innovate systems, so it will record wide band data, essentially removing the need to use a dyno to record AFR. There is a dataviewer which can be purchased separately, but is good to highlight a specific point in the map to concentrate on. I fitted the wide band sensor, and have the gauge in the front cup holder which fits perfectly.
Just my2p, you'd be welcome to sit in the car at some point to gain an idea, it isn't a race car, just fast road suppose, but 250chp with standard exhaust cat back, standard I/C, and standard cylinder head currently, the car looks standard in every way.
Silent1 said:
That sounds mighty tempting, does anyone know of mapping software available for the first generation BMW mini (r53) it's a chrysler tritec engine (IIRC)
I still intend on hopefully paying someone to walk me through it to begin with, i don't intend on just diving in and fking it up
You can use Bytetronik which is what Tom uses, however you only get the standard ~30 maps against the 700'odd that he will have access to. Be aware that the OEM R53 ECU isn't an easy one to begin your calibration with, not least because it's a MAF based system that uses 2 map sensors rather than a maf sensor.I still intend on hopefully paying someone to walk me through it to begin with, i don't intend on just diving in and f
king it up I admire your willingness to learn but in all honesty, having seen a few of Tom's tunes, he's quite good at it particularly on the stock ECU. Maybe he's not up there with Chris, but then again he doesn't have the same experience. I think a lot of your concerns about the power graphs are to do with the 469 Catcam which is very aggressive for road use
Old thread I know, however, if you want to learn ECU programming / tuning, I can highly recommend:
1. Booking the academy training course with Sam at TDi South - I found this really enjoyable and my first go and actually running a car on a dyno and mapping it myself
2. Checking out HP Academy (https://www.hpacademy.com), I have learnt a great deal for this and it really helped before going to the Academy day with Sam
Once you are ready, TDi actually hire out the hub dyno setup for £60 per hour and have had several people hire it simply to learn mapping on their own engine (Sam will step in and help out for chargeable 15 minute chunks) - one guy spent quite a few days, mapping the engine, wiping the map and starting again the next day until he had it refined.
I would say experience counts for a great deal, it is ONLY SCIENCE, but having done many engines and being familiar with the software sure makes it quicker and easier.
It does make me laugh, now I have an understanding, how people can have an opinion on how to map, or "map for torque" etc - utter bks... The engine wants what the engine wants, it's not a matter of opinion, it's fairly basic physics.
1. Booking the academy training course with Sam at TDi South - I found this really enjoyable and my first go and actually running a car on a dyno and mapping it myself
2. Checking out HP Academy (https://www.hpacademy.com), I have learnt a great deal for this and it really helped before going to the Academy day with Sam
Once you are ready, TDi actually hire out the hub dyno setup for £60 per hour and have had several people hire it simply to learn mapping on their own engine (Sam will step in and help out for chargeable 15 minute chunks) - one guy spent quite a few days, mapping the engine, wiping the map and starting again the next day until he had it refined.
I would say experience counts for a great deal, it is ONLY SCIENCE, but having done many engines and being familiar with the software sure makes it quicker and easier.
It does make me laugh, now I have an understanding, how people can have an opinion on how to map, or "map for torque" etc - utter b
ks... The engine wants what the engine wants, it's not a matter of opinion, it's fairly basic physics.Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff