Source for electronic components - injector driver
Turbo Mini Forums

Gents

I don't mind admitting I'm slightly out of my depth, but treading water furiously!

Trying to repair an 80's 5th injector controller for a 205 turbo technics.

I think I need to source a replacement L9336 which as far as I can tell is a peak & hold injector driver.

I've tried google and RS/Farnell but so far no good.

Any recommendations where I can look? I am sure the MegaSquirt savvy folk may be able to direct me.

Cheers

Might be worth an email
http://www.4starelectronics.com/part_detail/L9336.html


I also found this pdf http://www.datasheetarchive.com/dl/Scans-0...ns-00158885.pdf

Edited by gr4h4m on 10th Sep, 2015.

I run a supercharger and I don't care the TB is on the wrong side.
VEMS + 12 PSI + Liquid Intercooler = Small Bore FUN!

Isn't peak and hold done with another chip and various resistors on the board? Those control the transistor so that it peaks and the holds the current at a value set by the resistors? Is it just the transistor you need?

Ah, ok its a complete package in one.......

I think if these were still available in some form, the megasquirt lot would be all over them. I know Jean has a peak and hold driver board that use a peak and hold driver chip and a standard type three pin package transistor. I may have one of these boards built up somewhere, if I can find it, its yours if you can cobble the parts for your needs.

Cheers gents. I have the relevant datasheet.
I think the issue is that someone has fitted the wrong injector, a 4ohm impedance instead of the correct 16ohm. I think this has fried the injector driver. What driver chip do the MegaSquirt use?

Was going to say i doubt any purpouse built peak and hold driver ic's exist these days, modern microcontrollers are so fast all you need is a suitable generic igbt/fet and a shunt and the rest is done in software.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

16 ohm shouldnt need peak and hold tom a fet/igbt that can handle the current should do the job. Peak and hold drivers are required for low impedance injectors as the energise the injector untill the coil in the injector saturates then they drop the voltage to the coil to hold it open wothout burning it out. A 16 or so ohm injector should tollerate battery voltage across them constantly.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

I'm still a novice at this so only trying to piece together what I can from datasheet etc.
This control box dates from the mid 80's so I guess they were doing the best they could at the time!
The original documentation states that a standard 205 GTI 1.6 injector should be used, these were all 16 ohm.
From my datasheet there are two of these peak/hold injector drivers. One rated at 2.4/0.6A and one at 4.0/1.0A. My man maths deduced that the first one is intended for hi impedance 16ohm inj and the second for lo impedance 4ohm, but again I could be wrong. The board I have has the 2.4/0.6a.

Ive got to be ohnest i think there overkill for a 16 ohm injector, drive current at 14.4v will be just under 1 amp and the higher current one should be able to drive a low impedance injector. Any chance of a link to the circuit or datasheet?

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

you shouldnt need peak and hold with 16ohm injector
it will draw less than 1A anyway

Yes i moved to the darkside

Instagram @jdm_brett

Yeah the peak and holds are normally used for low impedance injectors. A lot of the early injection stuff use to be low impedance. Nowadays injectors are mostly high impedance with the exception of some super huge flow injectors.

Data sheet for the L9335/6 here

http://www.datasheetarchive.com/dlmain/Dat.../DSA-153974.pdf

I am still piecing together a diagram for the entire thing, but the basis is a 7556 timer chip, with one astable timer circuit feeding into another. The output from the second feeds a signal to the 9335 which fires the injector.
It's a basic setup, only 3 wires in, live, Earth, and pulse from the ignition coil neg. there's a 2 bar abs map sensor mounted on the case.
There are two adjustment pots on the board which I have yet to figure out, I think one is used to alter the point at which the timer chip starts to work (e.g. As soon as the map sensor sees boost), the second I think is part of a voltage divider circuit which in turn allows duty cycle from the timer to increase as boost increases. The frequency is set by the pulse input from the cool and so rises as revs rise.
All pretty Heath Robinson, as this is working alongside all the original injection system, but when it all is working right the cars do drive surprisingly well.

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

http://uk.farnell.com/texas-instruments/lm...-dip/dp/1564728

This is current but will require a drive darlington and a few other support components but should replicate the chip you need perfectly

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

Edited by Rod S on 11th Sep, 2015.

Schrödinger's cat - so which one am I ???

Gents

Firstly thank you all for your replies and information.

I spent a good few hours trawling the net for data sheets and figuring out what the circuit was doing. I don't profess to be any kind of electronics genius but I did study it at AS level some years ago and suprisingly found I remembered a reasonable amount.

Anyway the crux really was that although I now understood what I thought the circuitry was doing, with the equipment I had to hand I had no real way of testing it.

Driving the car up the road even with a wideband and boost gauge fitted was annoyingly still inconsistant.

Anyway so I called on the services of a mate of mine, he is an electrical engineer and has a keen interest in anything electrical, he is currently building his own 3D printer.
But most importantly he has an oscilloscope.

So went round with the control boxes, I had managed to acquire a second one, suprising given as these things are the equivalent of rocking horse poo.

The first challenge was making some kind of simulated signal, the input is from the neg side of the coil. Simon is a ingenious chap so he soon rigged up an Arduino programmed to give us a square wave of 33hz to emulate 1000rpm.
With the box powered up and the signal fed in we were able to follow the signal through various stages of signal processing.
The crux turned out to be a bad ground in both cases to the injector driver, for some reason instead of being mounted on the main board it was situated on its own little bit of prototype board.
With a bit of soldering all of a sudden my test injector plugged into the loom started making the feintest ticking if held right to your ear. This didn't seem right, and the power supply showed it not to be drawing any current. We decided to feed it direct power and ground to test it, and it then made a proper "click" as you would expect. Connecting it back to the control and now it was clicking away merrily as you would expect.
Once we got to this stage we put the scope onto the driver chip output, sure enough the peak and hold trace was clear to see.
I hear what you are all saying that modern injectors do not require it, but I can assure you it certainly is there! All I can put that down to is that in the mid 80's when they designed it, they did so perhaps to allow it to work with low impedance injectors. The Turbo Technics 5th injector box was also used I think on other vehicles that maybe used a low impedance injector at the time.

Anyway the upshot is two working boxes now, a very satisfying few hours really in being able to properly test and repair.

yeh

Good result!

Team www.sheepspeed.com Racing