Measuring soldered-in coil resistance with a current source.

[cupoftea]

Hi,
Is this SOIC8 NFET rated for the current it conducts in this 20A buck current source?
The FET conducts 20A high pulses at f(sw), but its average current is only some 2A…. Datasheet says it can handle only 11A pulses, so will it be alright?

NFET DMT6006LS

https://www.farnell.com/datasheets/2918042.pdf

 

[barry]

Data sheet says it can handle 110A pulses.

 

[crutschow]

Deleted

 

[cupoftea]

Data sheet says it can handle 110A pulses.

Thanks, yes, though thats at 1% duty...the buck duty would be more than that.

 

[cupoftea]

Hi,
We need to measure the very low resistance (~ few milliohms) of transformer coils when they are mounted into a PCB. This is to see whether or not they are properly soldered into the PCB. Its impossible with eg Triple insulated Litz wire, to be sure that solder pot dunking has stripped every strand, as well as the triple insulating jacket…..best way is to solder the coil in to PCB….then shove known current source ~20A through the coil, then measure voltage across coil….hence resistance.
The attached ~20A current source is I believe the cheapest way to do this. Would you agree?
Is there anything cheap offtheshelf that does this?

LTspice sim and PDF scm as attached
(Its duty cycled to reduce dissipation......555 time constant would be longer than seen here)
(Will use an IRS21271 for the bootstrap driver, and will just tie the CS input to hi-side GND))
https://www.infineon.com/dgdl/Infin...N.pdf?fileId=5546d462533600a4015356768e7a27c0

 

[KlausST]

Hi,

some thoughts on in circuit testing. You have to be sure not to test your application to death.
Testing with 20 A ... I see some issues:
* in "not connected" state there will be some higher DC voltage across the open test points of your measurement device. Be sure this "open circuit voltage" is rather low. I recommend less than 0.5V.
* in connected sate the current builds up to 20A. If you now disconnect there will be the energy stored in the coil that will be released and will cause very high voltage. So I guess it can easily high enough to destroy some circuit. Mind thath on manual testing the current direction applied to the coil may be random .. thus the voltage polarity will be random, too.
* Mind that there is a capacitor that keeps some energy. So when going from OPEN to CONNECTED state this energy will be going somewhere. Be sure it can´t harm any electronic parts.
* also mind ringing (resonance) when connecting and disconnecting.
* mind that on manual testing the wires of the tester may accidentally connect to random parts of your circuit... and 20A may burn a piece of copper trace.

Klaus

 

[FvM]

A standard lab supply sourcing 1A and a DVM with uV resolution gives me 1 microohm resistance resolution. Sufficient for most PCB trace and power magnetics DC resistance measurements.

 

[cupoftea]

DVM with uV resolution

Thanks, yes, i think cost would prevent us from doing that.