search anatech and jig and find the schematic for a two transistor matching jig with method.
If matching a pair is to produce any form of success then you MUST ensure that:
the temperatures are the same
the Vce is the same
the Ic is the same
the hFE is within 5%
the Vbe is the same
the transistors are coupled thermally and draught free.
Then you find if the two almost matching transistors track each other as you vary the Vbe. The Ic of both should remain very close to matching over a range of current either side of the operating current you have chosen.
If matching a pair is to produce any form of success then you MUST ensure that:
the temperatures are the same
the Vce is the same
the Ic is the same
the hFE is within 5%
the Vbe is the same
the transistors are coupled thermally and draught free.
Then you find if the two almost matching transistors track each other as you vary the Vbe. The Ic of both should remain very close to matching over a range of current either side of the operating current you have chosen.
AndrewT wrote:
Guys, there are few different schematics in web. One is with 2x26V transformer, other for 2x35V (50V on rails). I just wonder, if on the scheme with 50V rails I would put transformer 2x26V (~35V on rails) what would be changes (if any)? Is this is going to change transistors work point ? Let's say I don't care about heat. The most interesting things for me are:
1) sound (will it change? for better or worse?)
2) if any modifications in elements are needed
3) if with smaller voltage, amp will be able to handle low impedance loudspeakers
Thanks in advance for your comments
Guys, there are few different schematics in web. One is with 2x26V transformer, other for 2x35V (50V on rails). I just wonder, if on the scheme with 50V rails I would put transformer 2x26V (~35V on rails) what would be changes (if any)? Is this is going to change transistors work point ? Let's say I don't care about heat. The most interesting things for me are:
1) sound (will it change? for better or worse?)
2) if any modifications in elements are needed
3) if with smaller voltage, amp will be able to handle low impedance loudspeakers
Thanks in advance for your comments
the sound will probably not change, if you run the amplifier at the same gain and with the same signal input.
If you cannot check the currents and dissipations in each of the components, then you should NOT be considering changing anything from a proven working design.
The output stage and driver stage SOAR is very closely tied to PSU voltage and device temperature as well as load impedance.
You cannot ignore heat.
You cannot willy nilly change supply voltage.
The amplifier will be more able to drive lower impedance loads if powered from a lower voltage supply.
If you cannot check the currents and dissipations in each of the components, then you should NOT be considering changing anything from a proven working design.
The output stage and driver stage SOAR is very closely tied to PSU voltage and device temperature as well as load impedance.
You cannot ignore heat.
You cannot willy nilly change supply voltage.
The amplifier will be more able to drive lower impedance loads if powered from a lower voltage supply.
Hello everyone, this is my first post so I hope you will forgive me if it's rubbish. I am experimenting with different designs of class A amplifiers to see which I like best. I do not require large output power, 20 or 25 watts per channel is ample. So far I have built three, these being the Pass F5, the Jean Hiraga 20 watt and the JLH with dual supply rails.
I have compared these side by side to find which one I like the most and I have reached a conclusion but before I can be content I would like to ask you knowledgable people a question. Is it possible to construct a KSA50 with plus and minus 22volt supply rails, to make a KSA25 if so what changes to the circuit are required. Or are there any other suggestions for the ultimate 20 watt class A that I can try.
The power supply that will be used is constructed with two 300VA 18 0 18 torroids 50A bridge rectifiers 4X 100,000 40V electrolitics and dual capacitance multipliers. This seems to work well on the three amplifiers mentioned above.
Any help would be greatly appreciated.
Thanks.
Alan
I have compared these side by side to find which one I like the most and I have reached a conclusion but before I can be content I would like to ask you knowledgable people a question. Is it possible to construct a KSA50 with plus and minus 22volt supply rails, to make a KSA25 if so what changes to the circuit are required. Or are there any other suggestions for the ultimate 20 watt class A that I can try.
The power supply that will be used is constructed with two 300VA 18 0 18 torroids 50A bridge rectifiers 4X 100,000 40V electrolitics and dual capacitance multipliers. This seems to work well on the three amplifiers mentioned above.
Any help would be greatly appreciated.
Thanks.
Alan
I would like to hear your opinion on the three amps.
I also have the KSA-50 clone, a real Aleph 5,
Bobo Zen V5, Brian Bell Aleph 5, and also building a JLH, and Aleph X...The JLH and AX are not quite done yet, but the Krell edges out the real Aleph 5 by a hair .
I see no reason why a 25 watt KSA50 can't be made. You would have that much more room to crank up the bias up further before you reach dissipation limits
I also have the KSA-50 clone, a real Aleph 5,
Bobo Zen V5, Brian Bell Aleph 5, and also building a JLH, and Aleph X...The JLH and AX are not quite done yet, but the Krell edges out the real Aleph 5 by a hair .
I see no reason why a 25 watt KSA50 can't be made. You would have that much more room to crank up the bias up further before you reach dissipation limits
zeners need to be changed
Neychi is quite right. Before adjusting the rail voltages too much (up or down) one should recalculate the values for the zeners and their resistors to keep the operating point of the LTPs the same, it's easy, but should not be ignored. I'm surprised there isn't a spreadsheet by now...
Plan B is to install the constant current sources described much earlier in this thread. They make the exact front end voltages more or less irrelevant. I successfully ran with +/-10v rails using this strategy.
Stuart
Neychi is quite right. Before adjusting the rail voltages too much (up or down) one should recalculate the values for the zeners and their resistors to keep the operating point of the LTPs the same, it's easy, but should not be ignored. I'm surprised there isn't a spreadsheet by now...
Plan B is to install the constant current sources described much earlier in this thread. They make the exact front end voltages more or less irrelevant. I successfully ran with +/-10v rails using this strategy.
Stuart
Hi John,Thanks for the reply.
I'll give a quick answer as it's going off the Krell thread. All three amplifiers were built using good quality components. After a lot of listening this is what I think ( it's all subjective of course, input source and loudspeakers will also affect)
The F5 was my least favorite, it sounded " thin"and not very engauging I don't think I could listen to it for a long period of time.The Hiraga and JLH were both smoother and more musical, much easier to listen to, but I prefered the JLH, bass was firmer,midrange was open and detailed, female vocals outstanding. The wife liked it best as well. ( need I say more ? )
Alan
Hi Stuart,
can you please explain how I select the Zener voltage and resistance. I will be using +and - 22volt rails, I do not understand about getting the operating conditions for the long tailed pair correct. Also can you give more details of the CCS you mentioned. Sorry to appear so dim, but I am trying to learn.
Thanks for your help
Alan
I'll give a quick answer as it's going off the Krell thread. All three amplifiers were built using good quality components. After a lot of listening this is what I think ( it's all subjective of course, input source and loudspeakers will also affect)
The F5 was my least favorite, it sounded " thin"and not very engauging I don't think I could listen to it for a long period of time.The Hiraga and JLH were both smoother and more musical, much easier to listen to, but I prefered the JLH, bass was firmer,midrange was open and detailed, female vocals outstanding. The wife liked it best as well. ( need I say more ? )
Alan
Hi Stuart,
can you please explain how I select the Zener voltage and resistance. I will be using +and - 22volt rails, I do not understand about getting the operating conditions for the long tailed pair correct. Also can you give more details of the CCS you mentioned. Sorry to appear so dim, but I am trying to learn.
Thanks for your help
Alan
well...
...the CCS is too much to put in a single post, searching this thread will get you the results you need, but if a simple turnkey solution is your goal it's probably not the right answer anyway.
In a nutshell, the long tailed pairs are constant current driven, and the current source they run from can be made voltage independent, at which point, within the limits of the transistors involved the operation of the front end becomes voltage independent.
I'll find a schematic and let you know about the resistors and zeners shortly, first I have to run to the post office.
Stuart
...the CCS is too much to put in a single post, searching this thread will get you the results you need, but if a simple turnkey solution is your goal it's probably not the right answer anyway.
In a nutshell, the long tailed pairs are constant current driven, and the current source they run from can be made voltage independent, at which point, within the limits of the transistors involved the operation of the front end becomes voltage independent.
I'll find a schematic and let you know about the resistors and zeners shortly, first I have to run to the post office.
Stuart
numbers...
The zeners probably need to be about 1/2 - 2/3 the rail voltage, so say 14v, the zener 'bias' current in the original amp was about 6ma, so the resistor in series needs to be ((22V-14V)/0.006A) or 1333 ohms. Keeping the same LTP current (3 ma) the emitter resistor in the long tailed pair will be (14V/0.003A) or ~4700 ohms.
Don't fixate too much on the exact values here, if your rails end up being +/-24v no big deal, none of the values above would cause anything to break.
HTH
Stuart
The zeners probably need to be about 1/2 - 2/3 the rail voltage, so say 14v, the zener 'bias' current in the original amp was about 6ma, so the resistor in series needs to be ((22V-14V)/0.006A) or 1333 ohms. Keeping the same LTP current (3 ma) the emitter resistor in the long tailed pair will be (14V/0.003A) or ~4700 ohms.
Don't fixate too much on the exact values here, if your rails end up being +/-24v no big deal, none of the values above would cause anything to break.
HTH
Stuart
Krell clone on audiogon
Hello to all,
I bought the KSA-50 clone that recently sold on audiogon and I have some questions about it. The output transistors run hot, 165F degrees on the right channel and 185 degrees on the left. Also, one of the other transistors (has a small heat sink attached to it) is running at 200F. Are these temps acceptable / safe? Love the sound of this amp, just want to keep it running for a long time.
BTW, I'm a lifelong DIYer and have built many amps and speaker systems over the years. I was able to buy this amp for around the cost of the parts ($350) and couldn't resist.
Hello to all,
I bought the KSA-50 clone that recently sold on audiogon and I have some questions about it. The output transistors run hot, 165F degrees on the right channel and 185 degrees on the left. Also, one of the other transistors (has a small heat sink attached to it) is running at 200F. Are these temps acceptable / safe? Love the sound of this amp, just want to keep it running for a long time.
BTW, I'm a lifelong DIYer and have built many amps and speaker systems over the years. I was able to buy this amp for around the cost of the parts ($350) and couldn't resist.
Hot Krell clone
Thanks AndrewT,
I have no idea about Tj but I would guess it's at least 50degF higher. I'm going to check/adjust bias as the left channel is obviously running higher/hotter. Got to listen to it for a couple of hours last night but I put a fan on it (yuck) to play safe.
Here's a link to a member's amp that is using the same boards as mine:
http://img60.imageshack.us/i/dscn12304et.jpg/
I would like to know which schematic to use and the suggested bias for this design.
Thanks AndrewT,
I have no idea about Tj but I would guess it's at least 50degF higher. I'm going to check/adjust bias as the left channel is obviously running higher/hotter. Got to listen to it for a couple of hours last night but I put a fan on it (yuck) to play safe.
Here's a link to a member's amp that is using the same boards as mine:
http://img60.imageshack.us/i/dscn12304et.jpg/
I would like to know which schematic to use and the suggested bias for this design.
