Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

How can I build a CMOS current generator ?

Status
Not open for further replies.
The circles are voltages for the simulation. They can be pulses or steady voltages to represent a battery.
Do you have something built? If so post the circuit so we can see the whole picture.
 
I only have the existing system, unchanged now for several years. I can send sketches of that if you like.
But I might build the one you show though I don't have my Hameg where I'm living st present.

My plan now is just to add the zener as you & Alec suggest, to get a faster field decline.

Summarising, I have two probs at present. One, if I reduce the Time for "Fire", the solenoid doesn't Open the "mushroom" and I need to increase Time to about 4 ms minimum. It should act much, much faster. (The mushroom describes a small ferrite piece - the solenoid acts on the flat top, and pulls a "stalk" up within its cylinder housing, exposing a bleed hole to a neck in the stalk).
Two, it is slow to close, so too much fuel goes in. (Two)is what we are dealing with in this dialogue and I think you have solved it for me.
 
Last edited:
The hard part will be finding a zener that will handle the current. Most are only about 5 watts. You need to be able to handle about 1.25 amps (50% duty cycle) Lets just pick a zener of 24 volts. So 30 watts. That leads to 6 or so 4 volters in series. If we knew the inductance we could use a snubber - The RC like you drew.
 
Inductance: of course this varies with current but an experimental injector, claimed to be good, built in the Univ New South Wales was said to be "rather high" compared to the Bosch unit, and was 2.3 mH.

Maybe L ~2 mH would approximate our coil.
 
Hi RON

Re 83, if I wanted 36 V avalanche, and your nominal current was 1.25 A, giving 45 W dissipation, would I use (just as an example) 9 X 36 V zeners in parallel? If each zener was rated 5 W.
 
They don't share well in parallel, but are ok in series.
But thinking about your circuit there is a high and low current. If you always switch to the low current after the initial pulse you could design for that.
Or if there is always more off time than 4ms that would also help, but it sounds like you are right up against it.
 
I had been wondering if I could use a transistor voltage regulator instead of the Zeners. (I had been talking about 36 V but let's say 24 V for now). I see a L7824 holds volts from up to 40 V down to 24 V. Before we start worrying about the "40 V", my question is: How could this be hooked up?
If In is from Ground in my circuit, and Ground is to Ground, presumably it wouldn't work???
 
Last edited:
Alec @reply 79

Let's call R R1 and add an R2 to earth, so the base is fed from a voltage divider.
Then on "Fire", the harvested charge in the capacitor (at the high voltage from the previous switch-ff induction spike) accelerates the solenoid opening. Would that work?
 
Last edited:
@ RONV

OK - can you clear one thing up?

Talking about getting rid of the induced current spike on switch-off, as fast as poss to allow quick end to fuel delivery, does the Zener go between the "bottom" of the coil and ground (supposing a Zener DID work ok) ?
I think it should - but you have put it between the 24 V source and the coil. That's wrong, surely?

NB My problem is to shut off the fuel injection when it should end. In the injection unit, fuel pressure is about 7 tons per sq inch !! This causes very rapid response from the tiny piston. But to act, a bleed hole needs to be shut and that's what the solenoid does. So it has to discharge very fast to collapse the field.
 
Last edited:
Current Generator - change

current-gen.jpgI have just realised something. When the coil is switched off, its polarity reverses. Of course, that helps close the bleed, because its acting on a ferrite actuator. (Bosch, FIAT, Ferrari, Delphi are clever folk).

Please help with this bit (Reply 87) : The "lower" circuit is the Hold circuit.
I will drop the Hold current right down to range between about 1 and 3 A. What do I do??? (What resistor do I change?).
 
You should be able to adjust the 220 ohm pot.. But, You may have trouble keeping it open at low RPM.
 
You should be able to adjust the 220 ohm pot.

With my circuit, would that bring the HOLD current down as far as ~2 A ? Surely I would need to increase the fixed Resistance in parallel with the pot. And if so, to what?
 
Last edited:
Four years on, I'm here again. First, in my ignorance (as all suspected) the suggestion was to fit MOSFETS, nothing to do with CMOS, as you wondered. (The two terms had come up & I had stuck with the wrong one).

Next, I found a very low value for the injector Resistance. Its about 0.15 Ohm - exactly the value found in a Russian (?) workshop manual for this injector. Also, the Inductance is also low, about 0.15 mH.
We think (again as you collectively divined) about 5 Amp FIRE lifts the plate fast, and the HOLD can be much lower - exact value - not known.

We found (for whatever reason) we needed >15 V to make our circuit do its job. MOSFETS would (I am told) allow a 12 V input (why, I do not know).

But does anyone want to have a go at altering the circuit in Post 4, using MOSFET(S) ?

Thanks muchly

Malc
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top