Hi I don't have any PCBs for sale, but I have two different ones for free.
You can see one in post #2,383 and the other in post #1,557.
But I have to ask you to cover the shipping costs. Both are in green.
Regards
Myles,
This 10R resistor normally carries no dissipation. But, should there be some RFI around, or maybe you are testing it at 20KHz full power square wave, the dissipation will shoot up alarmingly. I normally use only a 1W resistor, but take some care to use a flameproof resistor and mount it 2mm above the pcb, so it won't fry it should it dissipate a lot of power.
Together with the 100nF series cap it loads up the output at high frequency, improves the stability of the amplifier in the >200KHz area where the phase shift CAN, in some situations, take the amp into instability.
HD
This 10R resistor normally carries no dissipation. But, should there be some RFI around, or maybe you are testing it at 20KHz full power square wave, the dissipation will shoot up alarmingly. I normally use only a 1W resistor, but take some care to use a flameproof resistor and mount it 2mm above the pcb, so it won't fry it should it dissipate a lot of power.
Together with the 100nF series cap it loads up the output at high frequency, improves the stability of the amplifier in the >200KHz area where the phase shift CAN, in some situations, take the amp into instability.
HD
Since there has been mixed pin outs on different transistors, I would like to confirm that I have the right pinouts. Please see pic below. I am using the Dacz pcb.
Q2 BD 139 is a given as it is on the heatsink and can only be mounted one way
Analyzing the schematic Q3 is mounted as ECB from the top to the bottom of KSA1381 as shown in the pic. (lettering face in to pcb)
I was able to magnify the pcb and check with schematic, and see that Q5 is mounted BCE reading from the left side of the KSA1381. (lettering faces out towards the inductor.).
Hugh Dean has assured that Q3 and Q5 do not need to be mounted on the heatsink and can be free air, If rails voltage is less than 45V.
If I have this correct, please let me know and I thank you for your assistance.
Q2 BD 139 is a given as it is on the heatsink and can only be mounted one way
Analyzing the schematic Q3 is mounted as ECB from the top to the bottom of KSA1381 as shown in the pic. (lettering face in to pcb)
I was able to magnify the pcb and check with schematic, and see that Q5 is mounted BCE reading from the left side of the KSA1381. (lettering faces out towards the inductor.).
Hugh Dean has assured that Q3 and Q5 do not need to be mounted on the heatsink and can be free air, If rails voltage is less than 45V.
If I have this correct, please let me know and I thank you for your assistance.
Thanks jvhb, I can see that the Q5 transistor shown is soldered BCE. The Q3 transistor, if orientated by the thick line is mounted EBC as I have stated above. I used schematics and looked at the traces on bottom of pcb.
I did not know about the thick line indicating back of transistor. Learn something new everyday
I did not know about the thick line indicating back of transistor. Learn something new everyday
Well yes, if you look at it backwards. But that can easily confuse things: An transistor with ECB pinout is (generally) always mounted "ECB"- when you view it from the front - where the text is printed
Myles,
Not critical. I would use around 10,000uF per rail per channel, should be enough.
As a rule, a total of 100uF/Watt per channel is sufficient. Increasing it just stresses the rectifier, which generates powerful spikes in the power supply and make it hard to reduce. So, for 42V rails, about 80W power, I'd used 8000-10000uF per channel, so that would be 4,700uF on each rail per channel.
Many argue about this and some use 200mF but that is very expensive and unnecessary.
HD
Not critical. I would use around 10,000uF per rail per channel, should be enough.
As a rule, a total of 100uF/Watt per channel is sufficient. Increasing it just stresses the rectifier, which generates powerful spikes in the power supply and make it hard to reduce. So, for 42V rails, about 80W power, I'd used 8000-10000uF per channel, so that would be 4,700uF on each rail per channel.
Many argue about this and some use 200mF but that is very expensive and unnecessary.
HD
Thanks Hugh,
This suprise's me, as I thought I would need about 40mF per channel (4 x 10,000uF on the CRC board). I will take your advice and use 4 X 4700 per CRC board/18.8mF per channel. Thanks for the explanation. May have to mix and match my supplies of 4700uF caps.
Edit: Hugh can you comment on my question in post # 2342/#2343 of the Alpha Nirvana 39W Class A thread. I ask about filter caps for the SLB psu for the AN. Thanks
MM
This suprise's me, as I thought I would need about 40mF per channel (4 x 10,000uF on the CRC board). I will take your advice and use 4 X 4700 per CRC board/18.8mF per channel. Thanks for the explanation. May have to mix and match my supplies of 4700uF caps.
Edit: Hugh can you comment on my question in post # 2342/#2343 of the Alpha Nirvana 39W Class A thread. I ask about filter caps for the SLB psu for the AN. Thanks
MM
Last edited:
Oops, what kind of circuit design is that? normal meander test and what happens?
Attachments
very tight mounting on such a board, with such an element base to do so closely is superfluous.
P.S. The Buche(Til) coil is wound very crookedly. there should be no resistor inside it. If according to the correct scheme, this coil is only to the detriment, it does not protect the NFB circuit in any way.
no problem, write exactly the supply voltage and current of the output stage, then I will write exactly.
I proceeded from my linearity parameters of this circuit, in which you will use it in a linear or curved way, I don’t know ...
the generated thermal power Q3 will be 0.25 watts, so a radiator needs to be put in... ...
the linearity of this circuit depends on the initial quiescent current of the output stage, the thermal power Q5 depends on this current. I repeat the question - what is the initial current of the output stage?