Hi
Some of you may have read my previous post, regarding the refurb of my two Maplin 100w Mosfet amp modules, where I was looking to Identify the Model of the Transformer.
Today I had another surprise to contend with.
Working on one module at a time, I have just finished replacing the electrolytic caps, and renewing the thermal compound, and Insulating washers under the Power Mosfets of the first module.
This done, I decided as a matter of course to check the Isolation between the Transistors and Heasink, and continuity between transistors where the connection is via a screw, all good there ... I then just probed around the transistors to make sure there were no shorts, and was surprised to see a reading of around 25R between the Source and Drain of the N Channel (2SK133) Fet, Checking this against the module I haven't started on yet the reading was around 600R.
I decided to remove the Fet and check it on the bench, just to be sure nothing on the board was causing this low reading, and again between D & S I still measured around 25R.
Being Familiar with the normal quick test by charging the gate to turn the fet on (Which I didn't really have to do in this state) I held my finger on the Gate and the resistance rose to around 1.2K, but drops back to around 25R when I remove my finger.
Any thoughts greatly appreciated.
BTW The modules were only handled at a static safe area, and once the fets were removed the case and legs were tied together with tinned copper wire, and stored in Static Shielding Bags.
Some of you may have read my previous post, regarding the refurb of my two Maplin 100w Mosfet amp modules, where I was looking to Identify the Model of the Transformer.
Today I had another surprise to contend with.
Working on one module at a time, I have just finished replacing the electrolytic caps, and renewing the thermal compound, and Insulating washers under the Power Mosfets of the first module.
This done, I decided as a matter of course to check the Isolation between the Transistors and Heasink, and continuity between transistors where the connection is via a screw, all good there ... I then just probed around the transistors to make sure there were no shorts, and was surprised to see a reading of around 25R between the Source and Drain of the N Channel (2SK133) Fet, Checking this against the module I haven't started on yet the reading was around 600R.
I decided to remove the Fet and check it on the bench, just to be sure nothing on the board was causing this low reading, and again between D & S I still measured around 25R.
Being Familiar with the normal quick test by charging the gate to turn the fet on (Which I didn't really have to do in this state) I held my finger on the Gate and the resistance rose to around 1.2K, but drops back to around 25R when I remove my finger.
Any thoughts greatly appreciated.
BTW The modules were only handled at a static safe area, and once the fets were removed the case and legs were tied together with tinned copper wire, and stored in Static Shielding Bags.
Short the gate to the source. Then measure the resistance from Drain to source. Watch the polarity on the ohm-meter so you don't end up measuring the body diode.
I would think that with the potential of the ohm-meter from drain to source, and with the gate shorted to source, that the ohm meter would read higher than 10K Ohms on a good device. WARNING: I haven't measured one...so this is just a guess.
I would think that with the potential of the ohm-meter from drain to source, and with the gate shorted to source, that the ohm meter would read higher than 10K Ohms on a good device. WARNING: I haven't measured one...so this is just a guess.
Thank you for your replies.
I'm a little wary of removing the Transistor for a third time, I'm not sure how the PCB solder pads will hold up, but I do need to verify if this transistor is ok to use.
When I originally noticed this problem and had the Transistor on the bench, when I touched the Gate pin I was actually connected to ground via a 10 meg resistor through the wrist strap, shorting the gate to ground (Or Drain) as you suggest may well give me a higher reading, so well worth a shot, Today I have been thinking about knocking up a simple test jig to test the theory, it maybe that the 25R I am measuring won't actually pass any current under a load greater than that of a DVM, Fet's can be funny things to measure, and I have had odd results in the past (many years ago) where the fet for all intents and purposes looked like a dud, but was actually fine (Will keep you posted)
Chris, The thermal paste I'm using is from RS (Radio Spares) and is the white type Silicone based type, and not the silver types used on todays CPU and coolers, I did read all of the general spec's, and it has a high Dielectric strength, also, all traces had been removed from the device while I was playing around with it on the bench, but good thought, and again, will keep you posted.
Best Regards.
I'm a little wary of removing the Transistor for a third time, I'm not sure how the PCB solder pads will hold up, but I do need to verify if this transistor is ok to use.
When I originally noticed this problem and had the Transistor on the bench, when I touched the Gate pin I was actually connected to ground via a 10 meg resistor through the wrist strap, shorting the gate to ground (Or Drain) as you suggest may well give me a higher reading, so well worth a shot, Today I have been thinking about knocking up a simple test jig to test the theory, it maybe that the 25R I am measuring won't actually pass any current under a load greater than that of a DVM, Fet's can be funny things to measure, and I have had odd results in the past (many years ago) where the fet for all intents and purposes looked like a dud, but was actually fine (Will keep you posted)
Chris, The thermal paste I'm using is from RS (Radio Spares) and is the white type Silicone based type, and not the silver types used on todays CPU and coolers, I did read all of the general spec's, and it has a high Dielectric strength, also, all traces had been removed from the device while I was playing around with it on the bench, but good thought, and again, will keep you posted.
Best Regards.
I get 4k on my Maplin mosfet amp.
You could do with a semiconductor analyser. It tells you gain etc
Ebay Manufacturer Refurbished - Atlas DCA - Semiconductor Component Analyser (DCA55) | eBay
I got my Maplin mosfet amp powered up after finding it buried in a box.
It mostly sounds ok but has quite a bit of hum.
The front end just has a resistor for a CCS and this allows hum off the supply in to the front end.
On my own redesign of the Maplin mosfet amp I added some decoupling to the front end and that sorted out the hum.
It mostly sounds ok but has quite a bit of hum.
The front end just has a resistor for a CCS and this allows hum off the supply in to the front end.
On my own redesign of the Maplin mosfet amp I added some decoupling to the front end and that sorted out the hum.
Not had chance of doing any further tests as yet, but if I can get a fairly decent reading with the Gate grounded, I might try and knock up a small rig whereby I can drive the fet with a square wave from my signal genny into something like a 500ma load, and see how it reacts.
Like I mentioned in the original post, holding my finger on the gate will let the resistance rise to around 1.2k, so I know the transistor isn't completely shorted, in the real world, and some real current to pass, it may well turn out to be ok ..... He said with his fingers crossed.
The other option would be to just power the amp up with a low voltage supply 12 ~ 15v rails (Limited of course) and see what happens !!
Like I mentioned in the original post, holding my finger on the gate will let the resistance rise to around 1.2k, so I know the transistor isn't completely shorted, in the real world, and some real current to pass, it may well turn out to be ok ..... He said with his fingers crossed.
The other option would be to just power the amp up with a low voltage supply 12 ~ 15v rails (Limited of course) and see what happens !!
Well, I have to admit I'm not sure whats going on with this Transistor.
I removed it again today, and did the normal Tests with the DVM and got the same results, shorting the gate to the source didn't appear to make any difference.
So with nothing really to lose, I connected it across a power supply set to 5v and limited to 100ma, in this state it drew about 17ua, placing a finger on the gate saw the current quickly rise to the 100ma limit, Connecting the Negative terminal to the gate briefly appeared to turn the device off completely (for a period of around 10 seconds) and then it would slowly start to conduct again.
Tomorrow, when I have some low rated fuses arrive, I'm going to connect it across a 15v - 0 - 15v supply, if it doesn't draw a silly amount of current and nothing gets hot or smokes, and I can actually see something like a waveform on the scope (not sure if I will at that voltage) then I'll try the main power supply.
Don't be surprised if a new thread appears asking if there is a kit available that rivals this little amp
I removed it again today, and did the normal Tests with the DVM and got the same results, shorting the gate to the source didn't appear to make any difference.
So with nothing really to lose, I connected it across a power supply set to 5v and limited to 100ma, in this state it drew about 17ua, placing a finger on the gate saw the current quickly rise to the 100ma limit, Connecting the Negative terminal to the gate briefly appeared to turn the device off completely (for a period of around 10 seconds) and then it would slowly start to conduct again.
Tomorrow, when I have some low rated fuses arrive, I'm going to connect it across a 15v - 0 - 15v supply, if it doesn't draw a silly amount of current and nothing gets hot or smokes, and I can actually see something like a waveform on the scope (not sure if I will at that voltage) then I'll try the main power supply.
Don't be surprised if a new thread appears asking if there is a kit available that rivals this little amp
Hi, Yes, there are some from Germany at the moment (2SK133) around £20.00 inc Postage and RS do the Magnatec version of the 2SJ50, not a matching pair I know, but it seems the 2SB48's are about as rare as hens teeth.
There is a post in another thread which mentions drop in replacements from Exicon? I think, but I'm not sure of their part numbers, and if they are indeed still available.
There is a post in another thread which mentions drop in replacements from Exicon? I think, but I'm not sure of their part numbers, and if they are indeed still available.
Last edited:
Not sure about your bare finger touching *very* high impedance gates test or what you intend to prove with that.
Try them in this simple gig which is designed for matching but as a side job will tell you bias voltage to pass about 11-14mA .
That alone confirms whether you have a functional MosFet or a blob of molten silicon, whether it´s a Lateral or Vertical MosFet, etc.
Here device is properly polarized, biased and connected; open gate or noise injection into an open gate mean little, if anything at all.
So measure what you have today and post results.
Try them in this simple gig which is designed for matching but as a side job will tell you bias voltage to pass about 11-14mA .
That alone confirms whether you have a functional MosFet or a blob of molten silicon, whether it´s a Lateral or Vertical MosFet, etc.
An externally hosted image should be here but it was not working when we last tested it.
Here device is properly polarized, biased and connected; open gate or noise injection into an open gate mean little, if anything at all.
So measure what you have today and post results.
Hi, Thanks for your reply, I am unable to open your attachment / Image so can't comment at the moment.
What I can tell you is that I powered up the module today on a 15v - 0 - 15v supply with the suspect Transistor in place, and the amp seems to work just fine, the only thing I did notice is a slight overshoot on the Rising Edges of a 1khz Square Wave, will have to wait and see how it compares to the other module once re - capped and the thermal compound replaced under the Mosfets, and of course on it's correct power supply.
What I can tell you is that I powered up the module today on a 15v - 0 - 15v supply with the suspect Transistor in place, and the amp seems to work just fine, the only thing I did notice is a slight overshoot on the Rising Edges of a 1khz Square Wave, will have to wait and see how it compares to the other module once re - capped and the thermal compound replaced under the Mosfets, and of course on it's correct power supply.
Posting here what I answered to your PM
To have + and - voltage, it is measured relative to ground.
There are no "single" voltages on their own, they are always measured relative to some point, in this case Ground.
Exact value will vary depending on individual transistor, and is exactly why we build and use this matcher in the first place: we get a batch of transistors, **they will be all different** because that´s a reality of life, we measure and label all, I suggest you attach a small sticky label to each, writing the value found, which is usually around 3.5V in VFets (IRFPxxx), around 0.6V in Lateral Fets (2SK/2SJxxx) , etc. and when building/repairing an amplifier you choose those with as-close-as-possible Vgs ensuring they will pass basically same current and share load equally.
No, R1 is a fixed 1k resistor.
That is Ground Symbol.
Yes.
To have + and - voltage, it is measured relative to ground.
There are no "single" voltages on their own, they are always measured relative to some point, in this case Ground.
Whe are measuring the Gate-to-Source voltage, also called Bias Voltage, needed to pass current around 1mA through the power MosFet.
Exact value will vary depending on individual transistor, and is exactly why we build and use this matcher in the first place: we get a batch of transistors, **they will be all different** because that´s a reality of life, we measure and label all, I suggest you attach a small sticky label to each, writing the value found, which is usually around 3.5V in VFets (IRFPxxx), around 0.6V in Lateral Fets (2SK/2SJxxx) , etc. and when building/repairing an amplifier you choose those with as-close-as-possible Vgs ensuring they will pass basically same current and share load equally.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.