This is what I would do, but only me...
Get a Scope, get Function Gen, Get a Dummy Load, may you can borrow it from a friend and of rent it for a day or two..
Connect all them to that amplifier,
This is a class A/B I guess so Bias should not be that high, also it's Mosfet there you can reduce " half "of it..
Max Power is 100Watts @8ohm this would then be 28.5 Volts RMS..
Make it 1 Watt 2.84Volts RMS Sine Wave 1khz make sure that you get a real beautiful Sine wave make sure that the SINE WAVE stays centered around 0 volts, meaning that the + and - Signal are even, then switch over to Square Wave, if the scope does not display OVERSHOOTS, your bias is set right. Now try to go all way up to these 100Watt if you like, no need all just 50 Watts is more than enough, and check out Sine and Square @ 1khz.. if there is no change then you are done.. I think a good bias would be something about 100ma Mosfets do not use that much BIAS, because it's voltage driven..
Since it's A/B have you checked the Mosfets that these match at least a bit so
J &K will work together...
Then check 20Khz Sine and also Square @1 watts that is all what it takes,,
and now check temp on the output transistors don't have them claim up more than 30 degrees Celsius when on idle, if these get up more then bias is to large or cooling is bad.. With this I mean the Heat Transfer from the Mosfet to the heat sink..
Also there do not USE THERMAL RUBBER PADS.. USE MICA Insulation Pads the thinnest you can get and use good heat transfer compound as well. I use Unick Silicone Heat Transfer Compound. But that's ONLY ME
I'm building Class A amps, so I can sing a song when it comes to cooling..
You know I'm not a Professional, but there are so many here on this site.
But till today of all these amps I have built, not one has given up life after the top cover was screwed on.. while building and testing there is always some smoke signs from resistors or transistors which mean to have to stop to work.. but not when top cover is closed. because then all what I wrote above has been tested that way..
I know you do not have access to that Tools I have, but an Analogue Voltmeter is more than enough to set the bias..
BTW a Bias set wrong will also be hear able.. sometimes..But for sure when set to low, the Amp will be not that responsive as when set right.. when set too high then the outputs will get hot even there is no signal which has to be amplified.. Here a small video about exactly this but it's testing my newest Class A... Look at it then you know what I mean.. Testing Class A Variable Current Source Amp - YouTube because of some experience on this site I chose not to use a Dummy load of which I have several, but choosing my I.Q. 4 Series Loudspeakers instead..
I think 100 to 150 ma there you will do no harm.. to the amplifier..
Hope could help you just a tiny bit..
Chris
Get a Scope, get Function Gen, Get a Dummy Load, may you can borrow it from a friend and of rent it for a day or two..
Connect all them to that amplifier,
This is a class A/B I guess so Bias should not be that high, also it's Mosfet there you can reduce " half "of it..
Max Power is 100Watts @8ohm this would then be 28.5 Volts RMS..
Make it 1 Watt 2.84Volts RMS Sine Wave 1khz make sure that you get a real beautiful Sine wave make sure that the SINE WAVE stays centered around 0 volts, meaning that the + and - Signal are even, then switch over to Square Wave, if the scope does not display OVERSHOOTS, your bias is set right. Now try to go all way up to these 100Watt if you like, no need all just 50 Watts is more than enough, and check out Sine and Square @ 1khz.. if there is no change then you are done.. I think a good bias would be something about 100ma Mosfets do not use that much BIAS, because it's voltage driven..
Since it's A/B have you checked the Mosfets that these match at least a bit so
J &K will work together...
Then check 20Khz Sine and also Square @1 watts that is all what it takes,,
and now check temp on the output transistors don't have them claim up more than 30 degrees Celsius when on idle, if these get up more then bias is to large or cooling is bad.. With this I mean the Heat Transfer from the Mosfet to the heat sink..
Also there do not USE THERMAL RUBBER PADS.. USE MICA Insulation Pads the thinnest you can get and use good heat transfer compound as well. I use Unick Silicone Heat Transfer Compound. But that's ONLY ME
I'm building Class A amps, so I can sing a song when it comes to cooling..
You know I'm not a Professional, but there are so many here on this site.
But till today of all these amps I have built, not one has given up life after the top cover was screwed on.. while building and testing there is always some smoke signs from resistors or transistors which mean to have to stop to work.. but not when top cover is closed. because then all what I wrote above has been tested that way..
I know you do not have access to that Tools I have, but an Analogue Voltmeter is more than enough to set the bias..
BTW a Bias set wrong will also be hear able.. sometimes..But for sure when set to low, the Amp will be not that responsive as when set right.. when set too high then the outputs will get hot even there is no signal which has to be amplified.. Here a small video about exactly this but it's testing my newest Class A... Look at it then you know what I mean.. Testing Class A Variable Current Source Amp - YouTube because of some experience on this site I chose not to use a Dummy load of which I have several, but choosing my I.Q. 4 Series Loudspeakers instead..
I think 100 to 150 ma there you will do no harm.. to the amplifier..
Hope could help you just a tiny bit..
Chris
Last edited:
I would say that you have set the bias and offset correctly. BTW Japanese transistors from the '70s almost never had the '2S' prefix printed on them, it was a given and just means transistor in the JIS (Japanese Industrial Standard) numbering scheme. The bias setting is specific to the actual transistors and you would expect it to need readjusting any time a transistor is changed. However I would have thought the actual amount of bias should be the same for the replacement Excon transistors and setting it higher just risks overheating and shortening the life of all of the components in the amplifier.
Not quite: 2S just means 3-pin silicon device(*) - could be a thyristor, BJT, JFET, MOSFET, triac, whatever. 2SA/2SB/2SC/2SD are the codes for BJTs, implying PNP or NPN, low or high frequency. 2SJ/2SK/3SK are FETs
There are 4-pin dual-gate FETs like the 3SK88 for instance. JIS semiconductor designation - Wikipedia
(*) Nominally the digit is supposed to be the number of pn-junctions, but this doesn't seem to been followed by anyone - triacs have 3 junctions for instance, MOSFETs have 2 until they conduct and then there's just 1 (whether its single gate or dual gate. PIN diodes have 2 junctions (but no pn-junctions!), yet I bet they have 1 as leading digit in JIS or JEDEC!
All of this is not only interesting, its slowly increasing my audio electronics knowledge base. This weekend I've been tied up in a project and getting ready for school opening. These days a total nightmare! So, I won't get to work on the amp till tomorrow morning. If all goes as hoped the bias and DC offsets will be checked and set, followed by just letting it play for a couple of hours. That'll go a long way towards gaining confidence in it and my work.
Finally got a chance this morning to get back to the AV5000. I had it playing channels 4 and 5 for about an hour and then checked channel 4's bias and DC offset. Initially the bias was 143mA and .01mV offset. I readjusted the bias to 150mA, since .01mV offset is within tolerance I left it alone. Then reassembled the board/heat sink unit back into the chassis and played it for another hour or so. It does sound great!
Hpro: the mosfets are running around 50 deg C. That's higher than the 30 deg C you recommended, do I need to lower the bias to lower the mosfet temps?
Hpro: the mosfets are running around 50 deg C. That's higher than the 30 deg C you recommended, do I need to lower the bias to lower the mosfet temps?
It's OK like it is.. keep it that way.
What was the original Pad between the Mosfet and the heat sink?
Happy for you that it worked out.
What was the original Pad between the Mosfet and the heat sink?
Happy for you that it worked out.
Morning, all!
This morning I received an anwer from Keith, Excon Tech, on my question about biasing their mosfets.
"150mA per MOSFET is ideal for the Exicon MOSFETs. They are temperature stable at that level and perform well as the amplifier is biased more into the class A range than with lower bias currents."
This should be very useful info for anybody who's thinking or uses Excon units in their amp.
Yes its the original sheet between the mosfet and the heat sink. I've never used silicon sheet interface, but I've read that as long as they aren't damaged and all the surfaces are clean its safe to reusing. Yes, I'm skeptical also!
This morning I received an anwer from Keith, Excon Tech, on my question about biasing their mosfets.
"150mA per MOSFET is ideal for the Exicon MOSFETs. They are temperature stable at that level and perform well as the amplifier is biased more into the class A range than with lower bias currents."
This should be very useful info for anybody who's thinking or uses Excon units in their amp.
Yes its the original sheet between the mosfet and the heat sink. I've never used silicon sheet interface, but I've read that as long as they aren't damaged and all the surfaces are clean its safe to reusing. Yes, I'm skeptical also!
It is not a good idea to reuse silicone heatsink pads. There are microscopic voids between a transistor and heatsink that impede heat transfer. Both heatsink paste and silicone pads are there to fill in the voids and exclude air, this to maximise thermal coupling. One compressed and thermally cycled, a silicone pad when reused will not properly fill the different voids between two surfaces, if not microscopically realigned as originally installed.
Sorry for not getting back, work kept me too busy. But gotta earn the audio money! LOL
I agree, I hated reusing the thermal sheets, but it didn't make much sense replacing them till I finished all the repairs. Usually I use Dow, 308 on desktop computer CPUs, and think maybe combined with a mica sheet might be a better heat transfer agent. Any suggestions of a good source? I'm sure its on Ebay, but I don't trust any of their vendors for this type of item. Maybe Parts Express, but these days they seem to be selling shipping with your items added!
I agree, I hated reusing the thermal sheets, but it didn't make much sense replacing them till I finished all the repairs. Usually I use Dow, 308 on desktop computer CPUs, and think maybe combined with a mica sheet might be a better heat transfer agent. Any suggestions of a good source? I'm sure its on Ebay, but I don't trust any of their vendors for this type of item. Maybe Parts Express, but these days they seem to be selling shipping with your items added!