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.

MOSFET use

Status
Not open for further replies.
Thanks Alec

First, yes, the 7812 gives ~10 v as you say. But the injector circuit is separate (12 v). The pin 4 is 4 v (or was today but its almost as if things change as I test. I think I have randomly done all that you say).

but I will have another go.

Whats maddening is that different bits have seemed OK but then later they do not.

Please look out for my reply tomorrow mid morn.

Malcolm
 
Last edited:
Whats maddening is that different bits have seemed OK but then later they do not.
In that case I'd be suspicious of any/all connections .....plugs/sockets/solder joints.
 
Hello Alec
Well, (after the messing around in a messy lab) there were some loose connexions but what I reported was a problem before (I had checke repeatedly).
So here is the answer to your question:

With 14.7 v in, and pins 1 & 2 connected to 0 v, pin 3 isolated, we get (in volts except pin 7)

1 > 0 5> 8.1
2> 0 6> 0.5
3> 4.8 7> 17 mv
4> 3.7 8> 12

Bad news is I have now been expelled from Lab because Health & Safety noticed.

Malc
 
From those readings it would seem either (a) there is a missing connection from pin 4 to the 7812 output, and/or (b) the 555 is faulty.
 
Thanks Alec

I'm not sure if I have the newly bought 555 in or the old one. But as I am "out of the lab", I can't check.
The connexion from the 7812 looks OK.

I think its a write-off. I simply haven't the tools, instinct with electronics, etc to continue.

We have a sound experimental motor and we can't Fire it because (since the original genius died) no-one seems to have the breadth of competence needed.

You rel simple circuit DID work tho control was poor because C5 missing.

(I also got a guy to build (money) a more complex version, Arduino etc, and it seemed to work on his bench but wouldn't work at the engine rig. I asked a former helper to check it out and its just languished.

Malc
 
Hello Alec
I am at a loss still.

On your circuit, I think all is OK. If I short to 0 v the Pin 2 lead (near the Optosensor) the light flashes and the injector works. We previously saw that the brief Rise at the Gate caused a square wave Drop at the Drain, and on the Hold FET, this could be varied with Pot 1. (about 1 > 4 ms)

But ever since I put in your C7, with the two R11, R12 (12k Ohm, not 10 k), a change has occurred. The optosensor RS 307913 no longer drops to zero, so the light does not flash. This was always reliable before. I no longer have access to a Oscilloscope. But with a good multimeter, we find:
a) 12.7 v supplied to the diode side, dropping to ~1.7 v when connected up. ie, presumed continuous infra red output.
b) 12.7 v at the transistor side, but not dropping to zero when reflection surface introduced. It drops to 5 or 6 V rather slowly (a few seconds) on the multimeter. And does not go to zero.
(If the transistor receptor output leads are shorted, of course the light flashes). This is true of three Optosensors.

I can't see why putting in your Pull up circuit would do this but it occurred immediately.

Should I try to drop the voltage to the transistor receptor?

Thanx

Malcolm
 
Just adding, if I bypass your C7 R11, R12 the Optosensor works, puts the light on, delivers power to Gate.
But there is no control over brevity, ie light stays on as long as reflection is present.
I need your RC timing but in some different way. I thought it might work taking Pin 2 to 0 v, via the C7 and a 12k Ω R. But that just holds the light on. How do I put a timer in? (A fixed short timer).

Thanks

Malc
 
From the datasheet for the optosensor, I see its transistor current may be anywhere from 0.2mA to 2.5mA (albeit for Vce=5V). A current as low as that would account for the 555 not triggering, because the 555's pin 2 doesn't get pulled low enough. Increasing R11 and R12 to 150k each, and reducing C7 to 1nF should fix it.
 
Hi Alec
Thank you.

I did the above and it made no obvious difference, ie the light went on if I shorted the Optosensor transistor. But not by reflecting the infra red.
I then (Own idea) lowered the voltage to the Optosensor transistor by half, and the light came on and stayed on. I reduced it by a third, and it was as if I had done nothing.

(Because you had reduced the cap) I shorted that out completely and, Lo, it seemed to work. The light went on with a reflection.
So I can get a response.

But it is timed by the reflection. The light is on as long as the reflector is aligned. So that is no good. It means current flows to the injector, not by my control, but by the reflection duration.

Do we need some pre-555 timer switch? Or does C7 ideally do that?
 
C7 and R12 define the pin 2 input pulse duration, but require a sharp-edged input from the optosensor. So, it looks as though your sensor transistor isn't switching fast enough. A pulse-sharpening circuit can overcome that. Do you have space on your pcb to accomodate, say, a CD40106 chip (it's a 14-pin DIP chip), or a second 555, or a transistor and the odd resistor or two?
 
Last edited:
Thank you, Alec
Do you have space
I would do anything !! Yes, I will do it somehow. What do i do? Buy this chip? You'd need to draw the circuit.
 
What we had originally was the voltage drop feeding a...
ic4093...
joining a ic4013, (pin3) which had a timer lapse RC between pin 1 (out) and pin 4, and 0 v.

Some people said this might be unreliable but I think it was OK. Is this similar to your CD40106?
Is the CD40106 better?

<Malcolm
 
Here's the suggested modified circuit, using two (which two is arbitrary) Schmitt trigger gates of a CD4093 quad gate 14-pin IC to make a monostable which generates a brief trigger pulse for the 555 even if the optosensor output varies slowly. Bear in mind that a CD4093 is static-sensitive, so take due precautions when handling/wiring it.
InjectorFETdriver5.JPG
Malc, yes each gate is an inverter, but use of two gates overcomes that.
I'd forgotten we had a 4013 circuit originally. As for reliability, I used to own an automatic transmission car where the engine speed was sensed using an IC from the CD4000 series. The IC never failed in the ten years I owned the car!
 
Last edited:
OK Thank you.
For an ignoramus, can you choose a layout (pins 1 - 14) and draw where they go? I presume 3a and 3b are pin 3 and pin 4 but not sure about other connexions.
You have shown various pins, but not all.
 
Last edited:
Ok. Here's the circuit with the pin numbers shown.
InjectorFETdriver6.JPG
 
I simplified and cropped the schematic:
 

Attachments

  • schematic.png
    schematic.png
    268.3 KB · Views: 134
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top