kick back diode selection

H

hax

New Member
Hi all,

I have a picaxe controlling a mosfet which in turn is controlling a 24V 3A DC solenoid.

I am after a suitable kick-back diode to use to protect the mosfet.

I am space constrained on the PCB and I wonder whether the 1N4004 diode will be OK in this application? I dont want to use anything bigger if I can help it.

The 1N4004 is only rated at one amp, will it be OK considering the load is 3A?

Of course the load is not going through the diode, the diode is only clamping the back EMF from the coil.... but I wonder if the back EMF needs a heftier diode. What are the formulas for working this out? Or a rule of thumb?

Better to use a fast acting diode or will a regular 1N4004 do?
 
E

evanh

Senior Member
Always best to overrate a diode. That said, such a short pulse even at 3 amps shouldn't be a worry for a 1 amp diode.


Evan
 
I

inglewoodpete

Senior Member
The 1N400x datasheet says that the diode can withstand a 30 amp impulse. The solenoid is 24v 3A, which gives a DC resistance of 8 ohms. This resistance alone will limit an impulse of 240v to 30 Amps. That's not the full answer, of course.
 
D

Dippy

Moderator
The back EMF will be proportional to several things including current, inductance, resistance and switching times.
I'm sure you can Google the equation from somewhere.

I would guess it's OK, but not sure. A quick look at the Data Sheet shows that the Fairchild version can handle 30A for 8mS non-repetitive i.e. not too often. Is that acceptable as I don't know your app/timings etc. I assume you missed that bit when you read the Data Sheet? :)

I'm using a 0.2mH / 4 Amp rated inductor , switching it at 20 usecs 50% duty from a 9V supply and, in my circuit, getting back EMF pulses (just a few microseconds) of over 100V into a high-ish impedance load. Things vary hugely as you vary the inductor 'charge' time.

If you have a 'scope and various diodes it's noticeable in some circuits. Try some Ultrafasts and Schottkys out. It's only pennies. Watch out for the voltage ratings with Schottkys.
 
E

evanh

Senior Member
You don't have to worry about fast recovery. It only becomes a issue when you are trying to turn the transistor on again while the diode is still conducting on the flyback. Ie: Like Dippy's fast modulated switcher.

And flyback current does not exceed the on-state current.


Evan
 
Last edited:
Y

ylp88

Senior Member
Most simply, the voltage generated is dependant on the inductance (L), the current fall (di) and the fall/collapse time (dt):

v = L di/dt

Assuming 240V can be modest! ;)

ylp88
 
B

BeanieBots

Moderator
It's fast turn on rather than recovery that is required. Several 100's of volts can be developed before the diode starts to conduct.
I would NOT recommend the 1N400X series for flyback except for small largely resistive loads such as relays.
Having said that, if it was all I had in the drawer and the product was not for a third party, I'd use one.

YLP88 has given the equation, however, without some good test gear and detailed spec for the solenoid, you are never going to get a value dI/dt to put in it.
If you have the time, patience and a good scope. Do some tests with a high voltage FET, inductor and an assortment of diodes. You'd be surprised at the difference a higher speed device can make to the flyback voltage.
 
D

Dippy

Moderator
Haxby, summary: try some, get a 'scope.

Without knowing the ins and outs of your circuit no-one can give you a definitive answer.
And if there is a little bit of sparklie not dealt with by the diode, then sensible track layout, capacitors, snubbers can prevent problems.

Get to it My Son!
 
W

Wrenow

Senior Member
For what it is worth, the 1N4001 is pretty commonly used in that application for the firing solenoids, but that is at 6-12v and I believe the amperage of the Clippard Mouse valves is somewhat les (of the top of my head, I am thinking 1.5A or less). My local Radio Shack was out, so I picked up what they had of 1N4003's and 1N4004's to put a new firing switch in my son's ship this week. Amd to drop the voltage a bit for one of the switches which is control voltage sensitive.

But, this is a new adbenture for me. Never done it before. My purpose is to replace his pneumatic firing servo with something a tad more reliable (i.e. an electronically controlled solenoid valve).

Cheers,

Wreno
 
E

evanh

Senior Member
I've thought of a possible source of problems. On the presumption of this solenoid being a linear sliding shaft type. There will be a much larger energising current required than that which holds it closed. If the circuit is turned off during this peak you could get much higher than 3 amps flyback current spike.

Some tests might be in order.


Evan
 
You must log in or register to reply here.
Top