DIY MOSFET tester

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I had a chronology of "Curve Tracer" articles in the microprocessor unit I described for AX about 10 years ago. There was a great article in "Ham Radio" describing a unit a fellow had developed to trace ham amplifier tubes.

I have since acquired a Tektronix 576, and have bought and sold via EBay auxiliary units made by Heath and Eico. I think tracers of this nature are good to demonstrate device properties, perhaps to a physics class, and to match devices, but to make your own SPICE models nothing beats dropping the data into Excel, or at least saving as a CSV file.

You have to talk nicely to the Tek 576. The CRT's are very difficult to replace and most have led lives of abuse. I don't think anyone has seen at 576 where the safety interlock hasn't been purposefully defeated!
 
Perhaps you can tell us what temperature coefficient those ceramic power resistor you use would have ? Also I see that those resistors are stack one above the other without air space in between. I wonder how hot they might get in continuous operation (rather than a single shot per minute or so).

Further more, I expect that at 40V 3A, those power transistors of the dynamic supply will dissipate quite a portion of that 120W. I wonder how hot do they get, again in continuous operation, on the aluminium heat sink that you showed in your photos.

Thx for sharing your progress. We have decided to make our own, write our own software, and design our hardware from scratch. Our version will eventually be able to support continuous operation at full power (40V 3A), and sufficiently dimensioned power MOSFETs (12x TO247 devices in total) on large finned heat sink, and will use reference resistors throughout, with temperature coefficient of 50ppm/K or less. These are loaded to maximum 30% of the rated power, and are all mounted on heat sinks in addition. The hardware will have its own µP and linked to the PC via USB (no RS232 required).


Best regards,
Patrick

PS For us, long term stability means at least one year between calibrations, and all curve tracers cross calibrated. Cross calibration is necessary as we are measuring devices at 4 different locations over 3 continents.

.
 
Last edited:
Perhaps you can tell us what temperature coefficient those ceramic power resistor you use would have ? Also I see that those resistors are stack one above the other without air space in between. I wonder how hot they might get in continuous operation (rather than a single shot per minute or so).

Further more, I expect that at 40V 3A, those power transistors of the dynamic supply will dissipate quite a portion of that 120W. I wonder how hot do they get, again in continuous operation, on the aluminium heat sink that you showed in your photos.

Thx for sharing your progress. We have decided to make our own, write our own software, and design our hardware from scratch. Our version will eventually be able to support continuous operation at full power (40V 3A), and sufficiently dimensioned power MOSFETs (12x TO247 devices in total) on large finned heat sink, and will use reference resistors throughout, with temperature coefficient of 50ppm/K or less. These are loaded to maximum 30% of the rated power, and are all mounted on heat sinks in addition. The hardware will have its own µP and linked to the PC via USB (no RS232 required).


Best regards,
Patrick

PS For us, long term stability means at least one year between calibrations, and all curve tracers cross calibrated. Cross calibration is necessary as we are measuring devices at 4 different locations over 3 continents.

.
The most temperature stability is ADC input resistor (91K/12K),
The ceramic power resistor is no important.

I hard to explain it with English,If you understand Chinese,then browse link
ÖÇÄÜͼʾÒǵڶþ°æÉè¼Æ½ø¶È - ÒôÏìDIYÂÛ̳ - HIFIDIYÂÛ̳ - Powered by Discuz!
of 235#
 
Yea, I agree, 6 years is long wait. I admire your patience. I once waited 6 minutes for something!

Anyways, who has actually built or used the MOSFET tester? I must also admit that I know little about the actual use of MOSFETs. I of course have hear about the open sounding analog power amp they are capable of.

Where I am heading with this is that I have (2) Yorkville EF508p which use MOSFET in their LF and HF biamps. One of the units is blowing the main fuses that fast! It turns out that it uses NOSFETs so I am not real sure about the aproper approach to test them w/o service notes, etc. So, I am looking for a tester that also handles MOSFET. I have found only two: the ELEKTOR discussed in this article and one currently on eBay made by Peak Atlas DCA for ~$80USD. Which one is recommended?

The Yorkville EF508p usesthese MOSFETs: 1RFP140N, 1RFP9140N, 1RFP23N50L. Some versions apply a 'crowbar' circuit in this porduct series. The IRFP's are very hard to locate! Right?

Suggestions please.

Thanks,



Should I purchase the
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.