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.

Electromagnet Control

Status
Not open for further replies.

willeng

Member
I have drawn a schematic that seems to work ok in LTSpice?
I found some IR2110 chips yesterday & have tried to use one & really I am not sure how I went except for the simulation.

I thought I may as well try to make it handle a bit of current & have a 24V source in the simulation, at 150A it seems ok, to me that is?.
I have attached a screenshot & the sim file.
Anyway, see how I went?

I lowered the Gate trigger frequency so the simulation runs faster, it's pretty slow on my computer at 25kHz.
There is also a few LTSpice Models I don't have so I just used what I had?
If any one has the Models I would like to get them?

Cheers
 

Attachments

  • Circuit-.PNG
    Circuit-.PNG
    42.4 KB · Views: 86
  • Circuit.asc
    8.8 KB · Views: 48
Last edited:

tcmtech

Banned
Most Helpful Member
I wouldn't w0rry about running the PWM frequency all that high. Given the huge amount of inductance that a electromagnet that size has a few dozen hertz is still plenty fast.
 

tcmtech

Banned
Most Helpful Member
To be honest I don't see why most any basic off the shelf low voltage speed controller like the ones used on smaller electric vehicles or battery powered equipment couldn't be used to do the job.

Most variable speed controllers from a common cordless drill or such tool are capable of driving much larger capacity switching devices that what are used in the tools without any modifications at all other than switching out the original power mosfets for a larger set.

Just my thoughts.
 

willeng

Member
Hi tcmtech,
John mentioned about the audible noise switching at lower frequencies which I understand & I thought was a good comment.
Ok, obviously the pwm frequency can be adjusted to suit all the applications the circuit is for, now & in the future.

As you say magnets of this type have a lot of Inductance, so do I treat this like an AC circuit where the inductive reactance XL is greater at higher frequencies.
Is this why you commented on this?

Yes, once again it's quite simple to go & buy something to do the job, but I am not going to learn much by doing that?

Cheers
 

willeng

Member
Hi Max,
Do you see a problem using the Mosfets, obviously I won't need the number of them I have in the Schematic as I was only testing to see how far I could push something without the need for a heatsink.
Just playing really.
I only require 12V & 75A so I will adjust things to suit.

The reason I wanted to use the IRF3205 is that I have about 50 of them to play with.

Cheers
 

tcmtech

Banned
Most Helpful Member
Hi tcmtech,
John mentioned about the audible noise switching at lower frequencies which I understand & I thought was a good comment.
Ok, obviously the pwm frequency can be adjusted to suit all the applications the circuit is for, now & in the future.

As you say magnets of this type have a lot of Inductance, so do I treat this like an AC circuit where the inductive reactance XL is greater at higher frequencies.
Is this why you commented on this?

Yes, once again it's quite simple to go & buy something to do the job, but I am not going to learn much by doing that?

Cheers

Sort of yes. Given their design and type of iron they are made from they have extremely high impedance to any AC power over a few Hz. Odds are you could put 120 VAC to your 12 volt magnet and you might not even get a audible hum from it.
I know the huge DC multi tens of KW magnets on the salvage yard cranes have impedances equivalent to several 10's of Henry's despite having a DC resistance of only 2 - 3 ohms cold.

What frequency you choose is up to you but if you go too high your own power lead impedances will start to come into play.

As far as simplicity and education go you sort of have to look at which is more important to the job at hand. As for simplicity a basic low voltage high power ~5 - 95% duty cycle PWM controller can be made from a 555 timer IC and large power mosfet using less than 10 total components.
 

ChrisP58

Well-Known Member
Most Helpful Member
I haven't run your simulation as I don't have a model for the IR2110. What do you see the current through D5 as? What is it's power dissipation?

The reason I'm asking is that, the current through the free wheeling diode will initially be the same as the peak current through the inductor and the mosfets. And, depending on the duty cycle of the pwm, the power dissipated in the diodes can be higher than that dissipated in the mosfets.

The free wheeling diode also needs to be a fast switching diode such as a schottky part.
 

tcmtech

Banned
Most Helpful Member
The free wheeling diode also needs to be a fast switching diode such as a schottky part.

This is an electromagnet built out of high loss soft iron alloy material that if it's having a really good day might reach tens of volts per millisecond voltage rise rates when the circuit is opened nothing like a ferrite core inductor in a SMPS that could have KV's/uS voltage spikes.

Any half assed power diode with a several tens of amps current rating would be more than sufficient for the free wheeling circuit.
Believe me compared to modern solid state power supply components these things have the electrical properties of steam engines not high end sports cars. :facepalm:

They may have gobs of power but they are slower than hell when it comes to changes in speed.
 

dr pepper

Well-Known Member
Most Helpful Member
With pwm its all about driving the fets correctly and getting them to switch quickly.

A flyback diode or r/c is enough to damp back emf on a lump of iron, tcm's right.
 

willeng

Member
tcmtech,
You mentioned about the Inductance -( Inductive Reactance) with the pwm at higher frequencies which I can understand, obviously with pwm you can't use XL =2 pi f L as with a sinewave.
So how do you calculate Inductive Reactance with a square wave of varying duty cycle?
Could you give me an example?

This may be the stupidest novice question ever asked on the internet?
If the pwm is running at say 20kHz as an example & 99% duty cycle, we have a lot of Inductive Reactance & then we go to 100% duty cycle which is essentially a DC output with no pulse & no Inductive Reactance, what happens when the impedance drops to near zero?
Obviously we have the DC resistance but does the current spike at the point when the Inductive Reactance drops?
Or do I have this totally wrong?

Cheers
 

alec_t

Well-Known Member
Most Helpful Member
If the pwm is running at say 20kHz as an example & 99% duty cycle, we have a lot of Inductive Reactance & then we go to 100% duty cycle which is essentially a DC output with no pulse & no Inductive Reactance, what happens when the impedance drops to near zero?
This sim shows the effect, on a 1H 2Ω inductor, of switching a 12V supply from 20kHz 50% duty-cycle PWM to 100% duty-cycle.
Electromagnet.PNG
 

willeng

Member
Hi alec_t,
A simulation is worth a thousand words, I can't thank you enough for that, it says it all.
Thank you again!

Cheers
 

jpanhalt

Well-Known Member
Most Helpful Member
It is not a 2Ω inductor. It draws 75A at 12V, according to the post, which is about 0.16Ω.

John
 

alec_t

Well-Known Member
Most Helpful Member
It is not a 2Ω inductor.
Agreed, but the sim was just to demonstrate an effect on an arbitrary inductor, rather than a practical circuit.
 

ChrisP58

Well-Known Member
Most Helpful Member
This is an electromagnet built out of high loss soft iron alloy material that if it's having a really good day might reach tens of volts per millisecond voltage rise rates when the circuit is opened nothing like a ferrite core inductor in a SMPS that could have KV's/uS voltage spikes.

Any half assed power diode with a several tens of amps current rating would be more than sufficient for the free wheeling circuit.
Believe me compared to modern solid state power supply components these things have the electrical properties of steam engines not high end sports cars. :facepalm:

They may have gobs of power but they are slower than hell when it comes to changes in speed.

I'm sure that you're right about that. My experience has been with high frequency core materials.

But I think you agree that the current required for the free wheeling diode in this application, will probably exceed what a little 1 amp rectifier can be expected to provide.
 

tcmtech

Banned
Most Helpful Member
But I think you agree that the current required for the free wheeling diode in this application, far exceeds what a little 1 amp rectifier can be expected to provide.

Yes and as I stated about using a power diode. I am not sure what constitutes a power diode in your terms but in mine it would be a stud mount or other such multi 10's of amps capacity type.

To be honest if a 1 - 2 ohm shunt resistor was placed in series with the free wheeling diode anything with a 20 amp rating would be more than sufficient given the equivalent series resistance of the coil plus resistor would limit the total peak amperage to well under 20% of the maximum running amps.
 

dr pepper

Well-Known Member
Most Helpful Member
Theres no reason why you couldnt use a low switching speed to compensate for high inductance, 50hz would probably work.
If the energy stored in the windings is high you could use something like a non dissapative snubber.
 

willeng

Member
Thanks to John's explanation of various things to consider in the datasheets & chrisP58 commenting on the diode across the inductor has led me to read quite a lot about mosfets & inductive loads.
As I have read it appears that a free wheeling diode across an inductive load using mosfets is not a good idea.
The freewheeling diode & the body diode of the mosfet form a very low impedance path which can cause high current destroying the mosfet if any reversal from the supply exists at any time which apparently does happen.
An active clamp seems to be the way to go, also the free wheeling diode makes the rate of decay from the inductor slow due to creating a lower voltage scenario across it.

I have read just enough to have more questions than answers so i'll ask one thing at a time?

In the graph below, I have the Zero Temperature Coefficient Point marked on the graph.
I have read that it is better to operate mosfets above this point because if you operate below this point with low gate biases the hotter the mosfet becomes the more current it draws & you can have a thermal runaway condition.
If you operate above this point, the hotter the mosfet becomes the less current it draws as noted on the graph.
Question 1 of 2000::)
Is this due to at low gate biases the mosfet is only partially turned on?



TRANSFER GRAPH.PNG


Cheers
 
Status
Not open for further replies.

Latest threads

Top