Hi Ian,
I think there's a point where speculation stops and experimentation begins. I think that there are enough good ideas on front end design already on the table. So I don't feel that much compulsion to unravel the UGS5 down to the last component. I have some ideas that I want to try that I think are within the boundaries of diminishing returns. We'll see. Again that's part of the fun of the thing.
The output transistor choice, however, makes a big difference. If these are 2sk1058/2sj162 family parts then the cost of building a prototype soars. These parts are several times more expensive than verticals plus there's more of them plus you need to buy lots of extras for matching. Serious commitment time. One needs to be pretty sure of a positive outcome.
I think that the TO-126 transistors are all the same part. They make up four current mirrors. The outboard transistors carry 5 times the current hence the heat sinks. IMHO.
Regards,
Graeme
I think there's a point where speculation stops and experimentation begins. I think that there are enough good ideas on front end design already on the table. So I don't feel that much compulsion to unravel the UGS5 down to the last component. I have some ideas that I want to try that I think are within the boundaries of diminishing returns. We'll see. Again that's part of the fun of the thing.
The output transistor choice, however, makes a big difference. If these are 2sk1058/2sj162 family parts then the cost of building a prototype soars. These parts are several times more expensive than verticals plus there's more of them plus you need to buy lots of extras for matching. Serious commitment time. One needs to be pretty sure of a positive outcome.
I think that the TO-126 transistors are all the same part. They make up four current mirrors. The outboard transistors carry 5 times the current hence the heat sinks. IMHO.
Regards,
Graeme
Hi Tazzz,
That is not my recollection. Here is a link to the posts I think you're referring to:
http://www.diyaudio.com/forums/showthread.php?postid=1232026#post1232026
Again the output side of a current mirror is a CS connection.
Regards,
Graeme
That is not my recollection. Here is a link to the posts I think you're referring to:
http://www.diyaudio.com/forums/showthread.php?postid=1232026#post1232026
Again the output side of a current mirror is a CS connection.
Regards,
Graeme
gl said:
I agree. There's plenty enough information already, and there
are plenty of ways to skin the cat. There's far more pleasure
(and bragging rights) in contributing some original decisions to
the project.
Also, the Laterals are a perfectly safe bet if you want to try them.
I would not expect much alteration to be required at all.
Hi,
I have trouble with my aleph X, i have build aleph v0.9 and v1 board and i have not solder J1a in both board.
I have 0.40ohm source resistor and i have trouble with offset i can't down it under 5V in both rail, i think y have correctly matched irf 9610 and irf044.
How i can check that matched transistor are good.
I can adjust pot for 0.5V accross source resistor.
I have 4 irf044 by board, my voltage is 20v by fet, i whant 2.5A curent bias.
Thank for your help
Sorry for my poor english
I have trouble with my aleph X, i have build aleph v0.9 and v1 board and i have not solder J1a in both board.
I have 0.40ohm source resistor and i have trouble with offset i can't down it under 5V in both rail, i think y have correctly matched irf 9610 and irf044.
How i can check that matched transistor are good.
I can adjust pot for 0.5V accross source resistor.
I have 4 irf044 by board, my voltage is 20v by fet, i whant 2.5A curent bias.
Thank for your help
Sorry for my poor english
noisefree said:
I appreciate that this is a relatively old post but I have been pondering the issue for some while. If Dirk's reasoning is correct, the front-end of an AX100.5 should apparently have a voltage gain of approx 32dB (26dB gain + 6dB feedback to double the natural damping factor of 75) but it seems to me that a typical UGS implementation has a gain closer to 60dB. I can't see any sensible way of lowering this to 32dB without 'throwing away' gain with a potential divider or similar and I can't believe this is done in the AX100.5. Is the reasoning here about the 32dB correct or is something else going on?
Ian.
60 dB is near to the maximum practical UGS open loop, and it's
pretty easy to drop down from that with degeneration or by
lowering the load impedances off the with resistor/semi
combinations forming current mirrors. Also it's popular to
load the input nodes with resistance to ground, which throws
away some of the open loop and buys better stability with
various sources. The net result is that you can have 6 dB of
feedback if you desire.
pretty easy to drop down from that with degeneration or by
lowering the load impedances off the with resistor/semi
combinations forming current mirrors. Also it's popular to
load the input nodes with resistance to ground, which throws
away some of the open loop and buys better stability with
various sources. The net result is that you can have 6 dB of
feedback if you desire.
Thank you Mr Pass for your reply. Perhaps I am being dumb but I’m not sure I completely understand all that you say. Presumably you are referring to degeneration at the diff pair sources? If so, doesn’t this reduce the SuSy effect as these will appear in series with the feedback? Does it matter?
There also seems to be a number of constraints in terms of reducing the resistors that form the current mirror, particularly if we wish to fix the current gain. In this case the ratio is fixed so both resistors need to be scaled down together. Won’t this reduce the gain of the diff pair (drain load) but increase the gain of the CS output? Perhaps the ratio is such that the gain of the diff pair drops more quickly than the CS increases?
I don’t understand your comment about loading the input nodes with resistance to ground. Do you mean right at the input as opposed to the virtual earth node? How does this have an effect? Also isn’t it also possible to reduce the gain by lowering the resistive load on the CS output? Is this a bad idea from another perspective?
Apologies for all the questions. When I have managed to acquire all the components I will try and find out some of these things myself.
Ian.
There also seems to be a number of constraints in terms of reducing the resistors that form the current mirror, particularly if we wish to fix the current gain. In this case the ratio is fixed so both resistors need to be scaled down together. Won’t this reduce the gain of the diff pair (drain load) but increase the gain of the CS output? Perhaps the ratio is such that the gain of the diff pair drops more quickly than the CS increases?
I don’t understand your comment about loading the input nodes with resistance to ground. Do you mean right at the input as opposed to the virtual earth node? How does this have an effect? Also isn’t it also possible to reduce the gain by lowering the resistive load on the CS output? Is this a bad idea from another perspective?
Apologies for all the questions. When I have managed to acquire all the components I will try and find out some of these things myself.
Ian.
Ian Macmillan said:
First off, Ian, the last time you were at my house you called me
Nelson.
Degeneration in the diff pairs will lower the Susy effect in the
same sense that it lowers the open loop gain - it takes more
differential input voltage to send the same amount of current
through the diff pair system.
I hope this is the right thread
Today I finished my first Aleph X.
I can adjust the bias nicely and can also adjust the relative DC offset to zero. BUT as soon as I connect the negative XLR input to ground (For RCA input) I get a relative DC input of 4V!
Absolute DC offset is 6V with XLR and 8V!!! negative XLR input grouded (for RCA input).
What did I wrong?
Today I finished my first Aleph X.
I can adjust the bias nicely and can also adjust the relative DC offset to zero. BUT as soon as I connect the negative XLR input to ground (For RCA input) I get a relative DC input of 4V!
Absolute DC offset is 6V with XLR and 8V!!! negative XLR input grouded (for RCA input).
What did I wrong?
I had the same problems myself using single ended inputs.Whatever I tried the dc offset at the output would be always very high and always drifting- very difficult to adjust.With balanced input there was never a real problem after adjustment and warm up things were quite stable despite initial high dc.I don´t know how Pass labs manage this in their amps.
When you connect one input to ground the absolute dc adjustment is feeding back to the relative adjustment and is almost impossible to adjust as it will slowly drift away.It is as frustrating as keeping water in your cupped hands.It will never stay.I think Nelson suggested lowering the output resistors to ground.
When you connect one input to ground the absolute dc adjustment is feeding back to the relative adjustment and is almost impossible to adjust as it will slowly drift away.It is as frustrating as keeping water in your cupped hands.It will never stay.I think Nelson suggested lowering the output resistors to ground.
