Just ordered the following for the GB300's:
C4, C8, C19, C9 and C5 with Blackgate N 10uf 50v
8 x Blackgate 100uf 100v for the bootstrap caps
For C3 I was thinking of something like a silver mica or tantalum?
What about the ceramics at C15, 16, 17 and 18 and the Wima cap in the center of the board, are these worth swapping out and if so, swap to what?
C4, C8, C19, C9 and C5 with Blackgate N 10uf 50v
8 x Blackgate 100uf 100v for the bootstrap caps
For C3 I was thinking of something like a silver mica or tantalum?
What about the ceramics at C15, 16, 17 and 18 and the Wima cap in the center of the board, are these worth swapping out and if so, swap to what?
ShinOBIWAN said:
Hi ShinOBIWAN/Macka
I think that C3 should be a silver mica, but, you have covered all the others. Wow, should be very nice tweaks, indeed.
Although, I have this feeling that C19 should be twice the size, ie. approx 100uF, or even larger 
The reason for this is electrolytic distortion. I can't remember the complete reason for this distortion, without doing some reading/researching. 
What do you think and where are you getting your Blackgates from I think that you should replace R2 with a very nice resistor, also.
rgds
I need help with matching
Hi SKA builders,
I need some help on the following issues:
I can not get the BC546C and BC556C but I have allot of the BC546B and BC556B.
I have written Greg and he suggests that I use the ones I have, but I should do the following:
“Measure hFE on as many BC546B and BC556B as you can and see if you can find
ones > 400 .
You can use these instead of C grade. Find 2 pairs of BC546 and 2 pairs of
BC556 - these will be your input pairs Q1,2 and Q3,4. They do not need to
match the opposite ones but the same types should be in pairs with a
variation of < 20 parts in 400”.
I got the following questions:
1.How do I measure hFE ?
2.How do I match the output mosfets ?
I have a DMM with the hFE funktion, but I just don't know how to go about it.
Thanks in advance for any help.
Klaus
Hi SKA builders,
I need some help on the following issues:
I can not get the BC546C and BC556C but I have allot of the BC546B and BC556B.
I have written Greg and he suggests that I use the ones I have, but I should do the following:
“Measure hFE on as many BC546B and BC556B as you can and see if you can find
ones > 400 .
You can use these instead of C grade. Find 2 pairs of BC546 and 2 pairs of
BC556 - these will be your input pairs Q1,2 and Q3,4. They do not need to
match the opposite ones but the same types should be in pairs with a
variation of < 20 parts in 400”.
I got the following questions:
1.How do I measure hFE ?
2.How do I match the output mosfets ?
I have a DMM with the hFE funktion, but I just don't know how to go about it.
Thanks in advance for any help.
Klaus
wwood said:
I would think that what you are seeing is the sig gen or
analyser residual distortion.
Sorry to say, but that is pretty poor performance for 5 watts /
8 ohms for a 150Watt SS design with global feedback.
Can you do a loop through test to check the equipment residual?
Cheers,
Terry
Hi Terry,
The data presented is accurate, and represents my unit.
1.) Both analyzer and generator have residual distortion more than 20 dB below measured distortion.
2.) The data is consistent with the designer's claims and his published data.
3.) The distortion drops smoothly with increased bias, dropping below 0.01% at about 100mA per device.
4.) Within the limits of my 150MHz scope, there is no sign oscillation.
5.) Why are you "sorry", it is what it is ?
Bill
The data presented is accurate, and represents my unit.
1.) Both analyzer and generator have residual distortion more than 20 dB below measured distortion.
2.) The data is consistent with the designer's claims and his published data.
3.) The distortion drops smoothly with increased bias, dropping below 0.01% at about 100mA per device.
4.) Within the limits of my 150MHz scope, there is no sign oscillation.
5.) Why are you "sorry", it is what it is ?
Bill
wwood said:
Hi Bill,
My post has a potentially negative slant and so I wanted to make
sure the measurements were accurate.
The spectrum is very good with a steep drop into the noise
floor as the harmonics rise.. You could say this is theoretically
textbook.
But the levels are relatively high for such a low power and into
8 ohms and 1k
The SKA acheives a lot of open loop gain with simple circuit. As such
it has the potential to very low measured distortion levels.
Is there any way you can push it a bit harder into 4 ohms and
higher frequencies?
I ask this as usually these are the measurements I do when
really looking for how a power amp performs. If I am shooting
for low THD, then I will try and give the amp a real stress test.
I am thinking with my designer hat on and trying to get a real feel
for this circuits technical strengths and weaknesses as you can be
sure that I am working on something that may better it.
As an example, a while back I was testing a power amp OP stage
and didn't have a lot of voltage level to drive it. So I loaded it with
2 ohms and drove 10kHz and 10V RMS to see what would really
unsettle it.
The meaured performance was in 0.00x % range with mainly low
harmonics.
Cheers,
Terry
Terry Demol said:
You do realise that the AKSA has around 0.05% distortion and sounds very good, ask destroyer X or pop into the tube forum if you fancy a heartattack
Distortion graphs aren't much more than ball park talking points and points of contention between more narrow minded individuals. A good amp they do not necessarily make and 0.01% THD with zero measureable higher order harmonics gets my 'it measures great' standard. Actually if Bill had come back and posted measurements with 0.1% distortion it wouldn't suddenly change what the amp sounds like.
Do designers actually listen to amp when you developing it or is it all scope and sketch work? Of course they do, don't they?
I have no problem believing that lower distortion is entirely possible but then what about the rest of the design? Like I said distortion is just an appendage to the whole picture. Looking for vanishing distortion isn't where your efforts should be concentrated. Clever and original design as well as great sound should be.
I've already said too much and I really don't want to see this thread turning into another chop it up and send it to texas job. If that happens I'll be keeping in touch with Macka, KLe, Bill and anyone else who want to share experiences via email as I just haven't got the time to run around in circles defending something that I really take great pleasure in listening too.
I think that Terry is simply trying to understand why SKA does what it does in his way. I bought SKA because I felt it was an interesting topology that solved a number of problems typically encountered in amp design and wanted to know more. I have done alot of measuring and some listening and agree with many in this thread that SKA is an excellent sounding amplifier no matter how it measures.
It measures how it measures and some of us will wonder why, speculate and make alot of measurements to try and learn as much as we can for our own designs. We each have our own philosophy of design and will pursue the best sound consistent with our philosophy. Like Terry I anticipated lower distortion at 5 watts than measured and that is a technical question to be answered for some of us just as others are exploring the sonic signature of component choices.
For me it would be enjoyable if this thread could find room for the subjective and engineering approaches together. Keep it respectful and we can all learn. I sure do not want to see this thread trashed again.
Bottom line is that I like listening to music through my SKA and I also wish to get a better understanding of why the amp measures the way it does, while respecting the designers IP. This is my hobby, not my profession and I stand to learn from all ways of looking at the amplifier from the talented folk at DIYA.
Bill
It measures how it measures and some of us will wonder why, speculate and make alot of measurements to try and learn as much as we can for our own designs. We each have our own philosophy of design and will pursue the best sound consistent with our philosophy. Like Terry I anticipated lower distortion at 5 watts than measured and that is a technical question to be answered for some of us just as others are exploring the sonic signature of component choices.
For me it would be enjoyable if this thread could find room for the subjective and engineering approaches together. Keep it respectful and we can all learn. I sure do not want to see this thread trashed again.
Bottom line is that I like listening to music through my SKA and I also wish to get a better understanding of why the amp measures the way it does, while respecting the designers IP. This is my hobby, not my profession and I stand to learn from all ways of looking at the amplifier from the talented folk at DIYA.
Bill
Does that mean that the politicians in Australia are the same as here in Germany?
By the way Hugh: thanks for helping me out with the hum problem of my AKSA the other day. I don't know why but I can not get access to the Audio Circle forum any more.
I still don't know how to measure the BC556\546. I lost my instruction manual for my DMM. Is here no one that can just tell me to put the red tip of the DMM to leg ? And the black to leg ?
By the way guys has anybody of you been able to compare the Rubycon ZL or ZLH caps in your amps with the Black Gates? There are some people who like them more than the BGs. I put the Rubycons every where, but I have no BGs to compare them with.
Klaus
By the way Hugh: thanks for helping me out with the hum problem of my AKSA the other day. I don't know why but I can not get access to the Audio Circle forum any more.
I still don't know how to measure the BC556\546. I lost my instruction manual for my DMM. Is here no one that can just tell me to put the red tip of the DMM to leg ? And the black to leg ?
By the way guys has anybody of you been able to compare the Rubycon ZL or ZLH caps in your amps with the Black Gates? There are some people who like them more than the BGs. I put the Rubycons every where, but I have no BGs to compare them with.
Klaus
Radian said:
Hi Radian,
I assume you are assembling a SKA and need to match the BC556/546 B/C transistors. I simply use the hfe function and pnp/npn socket on my multimeter. I read the datasheet to work out which is a pnp or npn transistor and which legs are b-c-e.
There are more elaborate methods for higher accuracy.
regards
Hi Radian,
the hFE function of the DMM must have a three hole socket for this purpose.
You disconnect the test leads and instead insert the 3legs of the BJT into the socket.
BUT you need to know which legs go where - C B E - it might be embossed in the case next to the socket. You need to know the orientation of your BJT legs.
the hFE function of the DMM must have a three hole socket for this purpose.
You disconnect the test leads and instead insert the 3legs of the BJT into the socket.
BUT you need to know which legs go where - C B E - it might be embossed in the case next to the socket. You need to know the orientation of your BJT legs.
AKSA said:I believe we have just witnessed the DIY version of the politician's NO COMMENT........
Hugh
I did have something more substantial but it was really just more fuel to the fire. I think I need to lighten up and perhaps not spend as much time on the forums and more time just listening to the amps.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.