Go Down

Topic: Coil Amperage Simulation and Actual Result mismatch (Read 871 times) previous topic - next topic

Tomas1337

Oct 27, 2015, 03:42 pm Last Edit: Oct 27, 2015, 03:44 pm by Tomas1337
Hi, All!

Hope you can help me with my small project. I have a coil, hooked up to a capacitor such as this


Coil = 0.0000137H
Capacitor  = 16V 2200uF
Power Supply = 10V DC

My coil is of AWG 31(0.227m diameter) wound about a 3m diameter, 8mm length core with 50 turns.
Calculations say I have about 0.0000137 H (From an online calculator). Putting this all into a simulations software(PSIM), it says that my coil current should reach about 60A in 500uS ( I assume that the software already takes into account all of the inductance and complexities of coils) . After the 500uS interval, I open my switch to cut the current off.

When I built my circuit, I hooked up a multimeter in series with the coil and am only reading a max of 2uA!. WTH.

So I guess my question is, what went wrong and how do I get more amps to my coil? (I need a temporary strong magnetic field lasting only 500us)

Thanks!

russellz

1. What's the value of the resistor?

2. A multimeter won't respond to 500 µS pulses very well, if at all.

3. I trust the wire diameter is 0.227 mm and the diameter 3 mm rather than metres!

4.  31 AWG wire will not stand 60 A.

Russell.
Retired after 40 years as a chartered engineer working mostly with RF and analogue electronics.

Tomas1337

Hey, thanks for the reply!

1. What's the value of the resistor? 0.
Removed it since I wasnt getting enough amps.

2. A multimeter won't respond to 500 µS pulses very well, if at all.
Ah really, Do you have any idea how to measure it? I dont have a ampere probe for an oscilloscope. I tried with a longer pulse of 1ms, but still the amps wont go even above 1A.

3. I trust the wire diameter is 0.227 mm and the diameter 3 mm rather than metres!.
Right. Sorry.

4.  31 AWG wire will not stand 60 A. Even for a 500us pulse every minute or so? It says that the fusing amps is 52A for 32ms so I was thinking of the deration of that.

russellz

You cannot measure current without some resistance in the circuit.  Connecting a multimeter will add resistance.  They mostly work by measuring the voltage across a low resistance.

To measure the peak current in a short pulse you can connect an oscilloscope (if you have one) across a low value resistance.  You could then insert the same value resistance in your simulation to compare.

The fusing current for a wire is defined for a straight wire.  The heating in a 50 turn coil will be much greater so the fusing current will be lower.

You will also need to think about the resistance of the piece of wire your coil is made from and add that to your simulation.

Russell.
Retired after 40 years as a chartered engineer working mostly with RF and analogue electronics.

Tomas1337

Hi, Russellz!

I did what you suggested and attached a 1.5Ohm resistor in series with the coil. Connected an oscilloscope in parallel with the resistor. I was getting a consistent 9.2V across the resistor. That means a solid 6A. But that doesnt mean that this 6A flows to the coil too right? In theory, it should be a lot higher, right? So how do i move on from here? How do I find the current passing through the coil and subsequently increase it?

Thanks,
Toams

jackrae

Hi, Russellz!

I did what you suggested and attached a 1.5Ohm resistor in series with the coil. Connected an oscilloscope in parallel with the resistor. I was getting a consistent 9.2V across the resistor. That means a solid 6A. But that doesnt mean that this 6A flows to the coil too right? In theory, it should be a lot higher, right? So how do i move on from here? How do I find the current passing through the coil and subsequently increase it?

Thanks,
Toams
Wrong !  If your resistor is in series with the coil then what flows through the resistor is also flowing through the coil.  I doubt you'll get 60A from a 2200microfarad capacitor and are you aware that an inductance endeavours to maintain steady state, so if the starting current is zero it will endeavour to keep it there.

Simulation software isn't all it's made out to be.

jremington

You can use LTSpice to simulate the circuit. Attached is an example, assuming the capacitor is initially charged to 10V and that the coil resistance is 0.1 Ohm. The current is unrealistic since the circuit does not include a value for the ESR of the capacitor.

russellz

Simulation software isn't all it's made out to be.
Like anything it's garbage in garbage out.  If you put the full circuit in, including all strays the results for a simple circuit like this will be spot on.

@Toams: Have a look at your simulation and try varying the value of the series resistor.  Remember also, as Jackrae suggests, the capacitor isn't perfect, it has series resistance (ESR) and inductance.  The coil will have series resistance and shunt capacitance although I suspect the capacitance can be ignored.

I'm not sure you have the coil right.  A simple single layer coil of 31 AWG with 50 turns will be more than 8 mm long.  A multilayer coil is much more difficult to design.

You will need to make the current sense resistor as small as possible.

Russell.
Retired after 40 years as a chartered engineer working mostly with RF and analogue electronics.

russellz

#8
Oct 27, 2015, 08:49 pm Last Edit: Oct 27, 2015, 08:51 pm by russellz
Just noticed that you have shown a voltage source across the capacitor so, in terms of the simulation the capacitor has no effect whatsoever.  In practice the connection from the voltage source will have series resistance and inductance and shunt capacitance so it will act as a transmission line.  The capacitor will reduce the effect that has.  I would assume that the capacitor will have an ESR of about 0.1 Ω and the coil has about the same resistance.

What sort of switch are you using.  That will also have a significant effect.

It's sometimes surprising how complicated a simple circuit can become :)

Russell.
Retired after 40 years as a chartered engineer working mostly with RF and analogue electronics.

Tomas1337

Hi, Russellz.

Thank you very much for the reply. My simulation has now become more realistic but when I did some changes on my actual circuit to match the simulation model, it still is a long way off. First off what I did was change my coil. Thanks for pointing out that my coil wasnt right. It was indeed multilayered and didnt know that it had such a big effect. I changed my coil to have 30 Turns in one layer.

I still added the 1.5 ohms in series as thats the smallest resistor I have of this moment. I will try and using a combination of parallel resistors to get it even smaller.

I have revised my simulation to be more accurate and here it is:


My coil still isnt getting anywhere the simulation result of 18amps but still only sucking in 6A. I want to get it much higher than 18A. Do you have a suggestion for me to do that? My goal is to create a strong enough magnetic field to reverse the polarity of an Alnico Magnet, which is the core of the coil.

Im using a MOSFET as the switch between the capacitor and coil.

russellz

With 10 V and a 1.5 ohm resistor you can't get more than 10/1.5 or 6.67 A.  You must use a much lower value of resistor to measure the current with your oscilloscope.  Something like 0.1 ohm.

You will probably also need to use several capacitors in parallel to get low enough ESR. 

Again, what are you using as a switch?  That must also have a very low on resistance.

Russell.

Retired after 40 years as a chartered engineer working mostly with RF and analogue electronics.

jackrae

If you can increase your supply voltage along with a higher voltage rated capacitor this will help in increasing current. You might be looking at several (or many) 10s of voltage.  The mosfet you use will have a turn-on period and a turn-on resistance, both of which will degrade the current pulse.  You might want to consider a shunt resistor to measure the current, say a 75mV drop at 100A  ie 0.00075ohm  (readily available from the likes of e-bay)

MarkT

it says that my coil current should reach about 60A in 500uS
And what material is the core, what MMF will cause it to saturate?  Cored inductors are
extremely non-ideal components, and if you overload them they become just a piece of wire,
because once all the magnetic material is magnetized the permeability drops to that of air.
You'll need to understand how to calculate magnetic flux density in a magnetic core and
compare that with the limit for the particular core material.  Core materials vary a lot in
what frequencies they can handle efficiently too, so it rapidly gets complicated.

You cannot measure a short pulse with a multimeter, you require an oscilloscope with a high-current shunt
to do this measurement.

Quote
After the 500uS interval, I open my switch to cut the current off.
And totally fry the switching device with the massive inductive spike I presume.  Trying to stop
60A flowing through an inductor is a mistake - you can only divert it or get a massive voltage spike.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

Tomas1337

You will probably also need to use several capacitors in parallel to get low enough ESR. 
Again, what are you using as a switch?  That must also have a very low on resistance.
It says on the rating that the ESR is only 0.03Ohms. Wont that be low enough since, 10V/0.03Ohms = 333A?

My switch is a MOSFET controlled by an Arduino. The specific MOSFET is TK62J60W.

TomGeorge

#14
Oct 29, 2015, 08:30 am Last Edit: Oct 29, 2015, 08:34 am by TomGeorge
Hi,
How do you know the exact value of your inductor? 0.0000137 H = 13.7uH

What is the dimension, wire size and turns of your coil.
Can you post a picture of it.


What is your application, why do you need a high magnetic pulse?
You will never get rid of all the series resistance, and caps, switches and inductors are not perfect in this world.

Tom.... :)
Everything runs on smoke, let the smoke out, it stops running....

Go Up