Power Flyback PCB. Other solution!!!

[GintsK]

So my tests on real engine was not so succesful:

1) My bench tests was using 30-40% PWM duty. In real life more than 75% needed. Because of this or another inductance of Audi 20V 5cyl n/a engines injectors heat stress was significant bigger. Things goes even worthier due 7Ohm IAC valve (what TVS diode do???). Here is additional aluminum wings as temporary solution:

2) Device last all break-in mileage and hard tuning sessions. But fails on second practice. On other region of my country. Thanks God it was best place to fail: diagnostic and repair make  one of the best car electronics expert here. My focus was on thermo load. He see a too high base voltage during transients as a reason. TIP107 emitter base voltage is rated 5V only. So he put MOSFET IRFZ46N in place. Gate voltage can be +/-20V. It survive rally-sprint race day later.

3)Circuit produce lot of noise. LCD shows scrawl immediately after start. Same man suggest put additional capacitors between flyback pad and power GND inside VEMS unit. So wires to external flyback unit no more works as noise source. In addition I solder 0.1uF cap on LCD between supply pads and shorten both - LCD and keyboard wires inside ECU. It works perfectly!


For less heat power and safer fail effect I decide rewire my unit back to INJ + and replace zener to 36V. 36+18V(max allowable U_batt by VEMS)=54V It is 6V below INJ FETs maximum.

More tests ahead.

Gints

[GintsK]

It works for me!
If four big injectors used cheap heatsink showed is more than enough, because engine can't be at so high power for a long period. In 400m bursts it stays cold.

PWMing is best way to control Low-Z injectors, but VEMS officialy didn't support high voltage flyback.
However anybody can try my solution which seems to work right even at hard loads up to 85% duty for 5 injectors (race proven)

I have 2L turbo project using 4x 1600cc. For first I try it with 30V TVS diode and  resistors from webshop. Idle was rough and no way run leaner than 0.85 at idle. Then I switch to PWMing and external high voltage flyback (you can try PWMing with same TVS diode at idle - it easily survive such low load). Results are fairly reasonable - idling at 1.0.

Gints

Transistor IRFZ46N or even IRFIZ48N (more robust) both cheap.
need mica insulator kit

Zener is up to 42V (because 18V protection diode together with this device gives 60V - FETs limit)

Resistor - 1kOhm

Note: device is connected in same way as common 30V TVS Zener diode. Only caps between flyback pad and power ground pad installed.

Gints

[[email protected]]

Awesome!  So are you running PWM and this flyback board?

[GintsK]

Awesome!  So are you running PWM and this flyback board?

Yes! And without any noise issues!

[MWfire]

What are your settings for injectors(open time,injpwm DC(%) and injpwm peak time)?

[GintsK]

For 1600cc i have
[email protected] - 784us
rampup  -          0
rampupbattfact - 4080 (disabled)
battfact    304 (probably too low)
DC            79.2%
peak time - 1.0 (probably can be less)

[mr_g]

MWfire and I did the active flyback for Low-Z injectors (1.7ohm).. I have tested it and it works great... The negative side of this one is that you need one for every low-z injector... But there is no heat dissipation..

[GintsK]

Can you explain little how it works?
As I understand it must work only at PWM time. Then high voltage spike must be allowed for fast injector closing.

what about noise? Did you use LCD?

Gints

[mr_g]

Yes. I have LCD... It's ok... This flyback is active one, so it does work only when injector is on PWM... So when injector closes transistor doesn't work any more and it act as there is no flyback... Just those 2 zenner diodes in series..

So, it's low voltage (14V) on PWM, and high voltage (45V) on closing injector... That's why it's called active flyback..

Injector ramp up time 0.9ms, PWM duty 30%... Injectors MED LPG 1.7ohm... They didn't wont to open on 4ohm resistor in serie... So they are big piggy injectors with 3A open current...