I have just had a revelation. Playing with the SERT programs has just started to make more sense.

For those who are interested...............

First I'm not looking for big numbers, just a bike that runs well, keeps heat to the minimum, reasonable mpg, and reasonable performance.

What have I done, well I loaded a canned map which is as close to my configuration as possible.
I reduced the rev limit by 200RPM which dosent sound like much but will make life quite a bit easier for the engine, and I very rarely hit the limiter anyway so it wont alter my riding style.

I left the idle revs just as they are, as I am concerned that oil circulation would suffer if I reduced it.

I increase the CLB tables to their max accross the board as suggested on this site, which has the effect of reducing the 14.6:1AFR in closed loop to about 14.1:1AFR, even though the tables still show the 14.6 to enable closed loop. Its kind of like tricking the system. And the advice I followed on here regarding this really did work, reducing felt heat considerably.

I then did a data run with my laptop, including the feedback from the O2. I want to keep the sensors operating as they will adjust the system sufficiently to allow for altitude and other minor variations, and I consider that to be a good thing.

On playing back the data I looked for excessive activity by the knock sensors which retard the timeing. When these are showing over 3% additional timeing retardation I start looking at the readings to see why, as it could be due to over high engine temps, lean condition, or of course too advanced timeing.

The O2 integrators will give a clue to lean/rich conditions as we are lookig for a 100% value to be ideal. If its higher that indicates a lean condition and if its lower a rich condition. I dont think its ever going to be consistently 100% and cannot see minor deviations mattering.

Now to get rid of the knock I can either retard the timeing(decrement or reduce the cell value)in that part of the map showing the problem or I can add fuel. Now since I need the AFR map to show 14.6:1 I cannot add fuel via the AFR tables, I've already maxed out the CLB tables so thats out as well. I now come to my revelation, which comes from finally understanding a little better how the various settings work together. I adjust the VE tables in the areas concerned. This tells the ECU that the cylinder concerned is taking in a larger volume of air, and so to maintain that 14.6:1(which is really already 14.1:1) it introduces a little more fuel, and makes everything a little richer.

I tend to find that the O2 integrators are not that far out so I only increase the VE cells by a max of 2 units(1%), and retard the timeing by a max of 2 units at the same time. These are small differences because as I said I am still learning and need to see the results. If I go large I may miss something.

Then it all starts again, I save the new map with comprehensive notes on what I have done, and reprogram the ECU. Then its another data run.

I know many of you guys will have known all this already but the effect the VE tables have on everything has only just fallen into place for me.

Paul

 

Thanks for that info Paul. I will be doing some data runs later today and may have some questions for you later as I start to learn the SERT a little better.

 

Roady,

I have found its best to do several short data runs to log different events, e.g. heavy through the gear acceleration, or constant cruise, or gently roll ons. You can do 4 or 5 runs while you are out, and try to make one of the runs under load like uphill which should really highlight any knocking issues. I have found that this way I can follow what I did on the run which helps.

I used the programs dyno simulator which while only an approximation, if you keep the setup info the same should show gains pro rata. So a 10% gain will still be a %10 gain even though the actual figures may not be spot on.

For example, yeaterday I did a run with the canned map(with clb maxed and rev set to 6000rpm) and got these results.

1st gear HP = 45.6 Torque = 54.8
2ndgear HP =71.5 Torque = 68.6
3rd gear HP = 74.8 Torque = 75.3

Today after alterations.2nd gear HP = 73.9 Torque = 73.8
3rd gear HP = 78 Torque = 79.2
And less knock retard activity

I just do very minor alterations so I can see the effects they have. Its going to be a lot slower than a dyno but a lot cheaper.

When I get the chance to get out early on a quiet straight bit of road I can check the dyno program through all the gears, but at the moment I am limited by safty limitations, and other road users[:@]

Good luck and let me know how you get on.

Paul

 

Loaded an updated map based on the data I recorded the last time out, only had to alter 4 cells on the rear VE table this time.

Bike runs strong, and no hotter than my old PCIII'd 88", low speed stumble pretty much gone, mpg still over 50mpg (UK gallons).

I dont think I will do much more to it, I'm pretty pleased.

Verdict............
If you have a stage 1 setup the SERT is pretty easy to tinker with to get a canned map to work. If you have a setup thats miles away from the canned map, I think a dyno and exhaust sniffer is the only way to go.

I will continue to do a few more data runs, but I cant see how I can improve on what I have much.

How did you get on roady

 

just for grins, what is your afr table for 20 kPa?

 

(deleted original question)

I have increased my CLB to max. All cells are reading about 798. Is this what you did to yours? How is your fuel mileage after your adjustment?

EDIT: haha, I guess it pays to RTFM. Explained everything Ineeded to know at the end of the closed loop explanation page. hehe

 

Hello-

Question: Once a SERT is loaded or has been used on a bike, can it be used on another bike?
I am not using mine -went to a Tmax.

Thanks,

Mark

 

No the SERT module marries itself to the bikes ecm when used for the first time.

WB

 

I see that you are adjusting the CLB values to get the AFR to 14.1, next question, can you then adjust the VE of each cylinder to get the AFR even richer to say 13.7 or 13.9 and still run in a closed loop condition? If so, does anyone know what VE number equals a point change in the AFR numbers?

 

The type of O2 sensor in use (the only type which can be with the ECU) only has a very narrow range over which it can linearly report AFR and 14.1 is about the limit on the rich end. Changing your VE tables will only result in a larger AFV (New VE) value and the ratio will still be 14.1:1. The only way to get richer is to forego closed-loop operation.