Turbo boost control box - reverse engineering

[TaDa]

Hi!

If I have stepped over the mark here on reverse engineering something I apologise now

But here goes...

A mate of mine showed me a little box on his car that he said allowed him to overboost his turbo.
Knowing I was interested in electronics he was hoping I could make him another to use on another vehicle.
I took a look and found it had 3 wires going into it - one went to earth, the other two were where an original wire going into the ECU had been split - so this was obviously messing with the signal coming from the turbo boost whatever - with no power it must be passive components only - I was feeling quite confident.
I only had a multimeter to hand and tried checking the resistances between the wires and this is where I became very confused.
The multimeter I was using had red and black leads and the red lead was positive.
If I describe the three wires as (E) Earth (S1) Signal1 and (S2) Signal2 I got these readings:

Plus Minus Resistance
E(+) S1(-) Infinite
S1(+) E(-) 25.12M
E(+) S2(-) Infinite
S2(+) E(-) 23.2M
S1(+) S2(-) 0.4ohms
S2(+) S1(-) 0 (effectively wire)

There was very little fluctuation on the measurements making me doubt there are any capacitors in there.
The E<->S1 and the E<->S2> can be described by a diode and a suitable resistor
but I don't really understand how the S1->S2 can be 0.4 but S2->S1 be 0!!

Can anyone shine a light on this?

 

[TaDa]

Ok - I've learned a little more.
It turns out that the original signal ranges from 0v to 5v (relative to the earth - well, this is a car)
If the voltage exceeds 4.5v, the ecu thinks the turbo is going to explode and sets the car into 'limp home mode' - ie turns off the turbo
Whilst the voltage is less than 4.5v, the ecu uses the information to dictate how much fuel is going to the engine.
So this box of tricks is basically allowing the 0-4.5v signals to go through unchanged
but not letting the voltage exceed 4.5v.

With passive components I can only see this being done with a zener diode - maybe with a few resistors to make it more graceful

But I still do not understand the Mega-ohmed resistances nor the weird flow on the s1<->s2 lines

Could there be a transistor in there? Yikes

So, any help still much appreciated

 

[alec_t]

Could there be a transistor in there?

Yes indeed. There could be all sorts of active and passive components. As for power for the active ones, that could be derived from the turbo 0-5V output, e.g. by using a diode and capacitor.

 

[TaDa]

Thanks alec_t.

I see how I could power active components with a capacitor/diode - but I'm pretty certain this wouldn't be needed as it simply needs to stop any signals exceeding 4.5v

Other than when the turbo was being pushed, the voltage would rarely exceed 3v - so unless there was voltage doubling in there too it would not have the power to generate necessary signals - I think this would make it over complicated.

The resistances between the S1/S2 and E suggest to me a zener diode and a normal diode in series to bump the zener's voltage breakdown a bit.
If I was testing with a 7v meter (It was a fluke - it said it took a 7v battery) then maybe the forward voltage was enough to breakdown the zener and allow some current to flow - hence the high resistance - but the reverse flow was completely blocked by the voltage dropping normal diode.

To my simplistic eye this makes the best sense of the signals to earth flows but the s1 to s2 weirdness still evades me - if there was an active circuit in there I'd have expected way more than a few ohms' difference.

Do I really need an active circuit to achieve this 4.5v signal capping?

 

[alec_t]

It's possible there's a FET in there. The drain-source resistance might measure differently depending on the polarity of your measuring probes.

 

[TaDa]

Hmm, I'll ponder on that thanks
Could an LM317 circuit demonstrate this too - if it had a diode on the adjustment line say?

 

[alec_t]

Could an LM317 circuit demonstrate this too

I doubt that would give you the measurements in post #1.

 

[TaDa]

Thanks again alec_t!

Can you think of a way to use a mosfet as a signal capping device in this case - where it is all driven by the power from the incoming signal?

 

[alec_t]

A mosfet with its d-s path in series with the signal from the turbo to the ECU could be normally on, but would progressively turn off as the turbo voltage rose above a threshold value.

 

[TaDa]

I'm out of my depth here - mosfets to me have always seemed like transistors on steroids.
From what I know (just read up on) they have 3 states - off, varying levels of on and then high resistance d-s.
They also seem to have high voltage requirements - whereas for most of the time I have less than 3v to play with - maxing at 4.5.
I fear this circuit is now beyond me - mosfets are too new to me and their characteristics so alien - and there was me hoping it was just a zener.
If you feel kind hearted and can give me a circuit to do this - it'll be appreciated but I think I'm going to have to throw the towel in now.

Cheers

 

[alec_t]

Personally I wouldn't build such a circuit. The turbo limiter is there for a very good reason! Also, a mod of this sort would probably invalidate your insurance (at least it would in the UK).

 

[TaDa]

Yeah - I hear where you're coming from but in this case the mod is legal - (you do have to tell your insurer about all mods - but this one is actually quite normal) - and though the limit is there for a reason, it is generally considered overly protective.

I'll ponder it a bit more - this really can't be too difficult - there are only three wires after all - and one of them is earth

Thanks for all your help alec_t

 

[debe]

When i reverse engineer something, the first thing to do is get to the circuit board. Then draw the circuit, not guess what its doing.

 

[TaDa]

Yes, that would be the best way to go but on this occasion not an option.

First the box was full of epoxy resin - there was no way of getting to the underlying circuit board
(and i believe it is a common trick to deface all the components before the epoxying anyway)

Secondly the box of tricks cost >£50 - thats a lot of money should the investigation break it.

No, on this occasion the only way is deduction.

One day I will get around to looking at this again - I'll tool myself up better and analyse it properly.
I'll check what current flows into the circuit as I slowly ramp up the voltage (to see if a capacitor is being 'filled') whilst monitoring the output.

I'm almost certain this is a simple capping circuit - and £50 is just a significant profit margin to offset the small market - just business in action!