Logarithmic TPS any experiences?

[matt_gsxr]

Hi there,

I was wondering whether anyone had any success with logarithmic TPS. It seems to make sense on some installs (I have gsxr600 ITB's on a GSXR1100) and low opening are a bit messy, and AE isn't the best.

Just wondering if anyone had any experience. I found that Webcon can supply these (see for example http://www.webcon.co.uk/Downloads/ACM-4.0a-S.pdf ) I don't know the price, which will no doubt be high. Any other vehicles fitted with them (I know Ferrari F40's are, but there aren't any of those at my local breakers at the moment).

Thanks,

Matt

[matt_gsxr]

So, I was looking at the TPS output (using the excellent PalmLog software).

It hangs around at the lowest ADC setting for a long time. I can get the engine up to 3500rpm before it increments (from 58 to 59, maximum is 225). Actually the car works nicely, but I think logarithmic TPS might help my acceleration enrichment, as I presently run this off the TPS dot, and it helps too little at small throttle openings, and gives too much at high throttle openings. I tried it with MAP. but that doesn't do enough just off cruise. I only have MS1, so can't blend TPS dot and MAP dot. I could invest in a daughter board I suppose.

To be honest it all works very nicely, but I am in search of perfection.


Matt

[PSIG]

I'm not sure what you mean - but I am assuming 'messy' means it's too sensitive? If so, you should be able to tune around it with larger TPSdot numbers, if they are appropriate. In effect, a small throttle change makes a rapid change in air flow, and so there should also be a rapid change in AE. This is normal. You may consider modifying to a progressive-angle throttle linkage to tame the throttle response (butterfly and MAP change) at low throttle, retaining with stock TPS. Do I misunderstand your issue?

David

[matt_gsxr]

David,

Sorry, poor wording.

By "messy" I meant, that the TPS ADC is not really representative of the amount of air going into the engine, and hence I have limited software control at these low throttle openings . Consequently the AFR is not as stable as I would like it to be. It isn't noisy (i.e. it is stable)

Really the TPS ADC is not sensitive enough for small throttle openings. Hence a logarithmic potentiometer (actually anti-logarithmic) that would give big changes in ADC for small openings, and smaller increments in ADC once the throttle has been opened. This seems to be available on after-market stuff (i.e. alpha webcon gear).

In my install at present I need a very small TPSdot values at low TPS to get decent AE. Whereas for higher TPS the TPSdot values that I see are very high, but the amount of AE needed is much smaller. I know there are fixes for this in later hardware (i.e. blended MAPdot TPSdot) but I don't have that.

I already have a progressive throttle linkage (it is standard on the GSXR600 ITB's, as the cable pulls on the throttle pivot which is elliptical and so does what you describe), and good throttle feel. The problem is that the ECU doesn't see enough resolution down at small openings.

Matt

[PSIG]

I see. Although I have tuned an ITB car, I did not have the issues you are experiencing. Then it would seem your solution has potential (no pun intended ). Alternatives would be a progressive (recessive?) TPS linkage creating the same effect, installing MAF in the air box inlet, or stepping to MS2, for examples. There are a number of 'squirters using these throttle bodies on various projects (especially as retro car ITBs) without issues - or have found simple solutions. Perhaps poking around for some threads on those on the and the other MS forums would be productive.

David

[matt_gsxr]

Thanks David for your thoughts on this.

FWIW:

I discovered a few things:

1) The gsxr600 TPS is not spring loaded, and although it doesn't appear to have any slop, the later TPS that Suzuki used were spring loaded. I have now used one of those.

2) A simple way of making the TPS non-linear (more effect at low openings) is to add a resistor between 5v and the wiper. In my case I have used 3.25k. By also rotating the TPS so that closed throttle gives 5 on the ADC, then I amplify the sensitivity at low throttle openings by around 2.5x. This was what I wanted to do (the www.locostbuilders.co.uk forum folk were excellent in providing this suggestion). There are already appropriate holes in the PCB for this resistor so I guess this is old ground than I am treading!

As stated before my engine runs and accelerates nicely already, I just think it can do better. Further, I think it is fair to say that the dotTPS approach is limited (hence why MS2 changes it). This non-linear tweak adds an additional bit of control for the cost of a single resistor. I think it should be standard practice with ITB's.


Matt

[PSIG]

2) A simple way of making the TPS non-linear (more effect at low openings) is to add a resistor between 5v and the wiper. ...

That certainly is simple, and a "Why didn't I think of that?" solution. Awesome! And thanks for reporting back with what you found so others can benefit.

David

[GeoffT]

I'm also using GSXR itb's, and have experienced all the small throttle issues mentioned in this thread, will definitely be trying the resistor between +5v and slider, sounds like a simple answer to one of the problems.

Another issue I've found is a more mechanical one though, in that when the throttle butterflies are fully shut on the overrun (or even at tickover) the high vacuum behind the butterflies seems to really grab hold of the spindles. This means you have to exert a fairly large pull on the cam to get them moving off the stop - then of course the vacuum lets go and an intended 1% rise becomes a 5% yank.....

It's almost like you need extra mechanical advantage for the first few percent of movement - the eccentric cam that operates the butterflies goes some way towards this, but it needs still more leverage to get off the stop. Spent the last half hour trying to design some sort of linkage to do this, now my head hurts so I'm open to suggestions.....

Good to see that I'm not the only one having hard to cure 'tip in' and ae issues though, starting to wonder if upgrading to MS2 might ease these problems?

Geoff.

.....interesting to see that 'j-e-r-k' got converted to 'yank' when posted - didn't realise that j-e-r-k was such a touchy word.......

[stevemgb]

I had a similar problem with AE on my old car and turned to RPM based AE.This transformed AE events but as my car only revs to 5500 it may not help you.Steve..