For those who may be interested. Excellent deal on R Cruz' suggestion of MR coupling.
The Madisound Speaker Store
The Madisound Speaker Store
I've just tested my PSU built with the GB PCB and using Hammond 229D34 transformers with 2x17V secondaries. My mains voltage is exactly 230V but for some reason (the 15% regulation perhaps) the transformers put out 39VAC which ends up as 52.5VDC unloaded at the DC outputs. Is that voltage safe to feed the Folded with or do I need to lower it and if so what would be the easiest way to do that?
Hello,
The board are ready.
Biaising done.
I tried, but I get an awfull "bzzzz" and almost no music. Any idea?
It is a high MM version:
The board are ready.
Biaising done.
I tried, but I get an awfull "bzzzz" and almost no music. Any idea?
It is a high MM version:
An externally hosted image should be here but it was not working when we last tested it.
Use 500 Ohm 10W resistor (or any combo like 1k//1k 10W) as dummy load so to create ~100mA draw to see how low your raw DC will go when loaded before you apply the real load (phono). Better safe than sorry. That way you can adjust the dropping resistors on the PSU board higher if the raw level stays uncomfortably high. Don't touch the dummy load and don't lay it against plastic surface etc. its gonna go hot.
No, its weak for constant dissipation at 100mA. Any extra 2SC5171 that you did not use in the KM maybe?
Something wrong in the signal side with a strong ground loop, or the bias is off in the stages, or the power side has a problem, I would look for. Can you post a photo of the build, and data about various points voltages in respect to ground? RAIL-GND, TP1-GND, TP2-GND, Q4 drain-GND, Q6 drain-GND. Are both sinks rising temperature?
Member
Joined 2009
Paid Member
I think you're on the right path. I think a bit of inductance is needed for a lively sound and the cap multiplier is the opposite. My reasons are related to work I've done on power supplies for a SET amp but I think they might apply here.
I have some of those
Can you direct me to the correct schematic ?
IYO what should be the major difference between this simpler CM relative to the Kmult ?
I was seduced by the Kmult because of it's very low Zout into 300mA.
Finally connected to power , and something is wrong
I did not bother powering the "other" channel - since I made them side by side whatever I did wrong on one is the same with the other.
Measurements:
B+ connected to the board at input - 50V
Rail to ground - 31.68V but never settles down
TP1 to TP2 - 25.25 more stable than the above - and maybe if left on longer it would (left on 10 minutes then turned off for fear I was harming some other component).
TP1 to ground - 25.4 same comments as above
TP2 to ground - did much bouncing around from 290 mV to 700 mV and then settled (rather suddenly) to zero.
The group pf four LEDs do not seem lit. They could be barely on or reflecting from the others (not sure - never used LEDs before)
The reg LEDs just might be too bright - but as above I am not sure.
The heatsinks never got beyond warm and felt equal in temperature.
I ended up using a 12R resistor in the LINK position. There was no heating of the power transformer. (Hammond 36V /115V - 2.8A)
Can you surmise from this what is happening?
Thanks,
I did not bother powering the "other" channel - since I made them side by side whatever I did wrong on one is the same with the other.
Measurements:
B+ connected to the board at input - 50V
Rail to ground - 31.68V but never settles down
TP1 to TP2 - 25.25 more stable than the above - and maybe if left on longer it would (left on 10 minutes then turned off for fear I was harming some other component).
TP1 to ground - 25.4 same comments as above
TP2 to ground - did much bouncing around from 290 mV to 700 mV and then settled (rather suddenly) to zero.
The group pf four LEDs do not seem lit. They could be barely on or reflecting from the others (not sure - never used LEDs before)
The reg LEDs just might be too bright - but as above I am not sure.
The heatsinks never got beyond warm and felt equal in temperature.
I ended up using a 12R resistor in the LINK position. There was no heating of the power transformer. (Hammond 36V /115V - 2.8A)
Can you surmise from this what is happening?
Thanks,
This answer is too simplistic.
A 100mA transistor can pass a continuous 100mA.
One simply has to ensure that the
is still within the limits of the device.
eg.
a 600mW device could have an arbitrary limit 300mW of constant dissipation set for the operational conditions and still be able to pass 100mA, continuously.
The manufacturer does not insert lies into the datasheet.
Yes, they can deliberately omit information to make their product look good. But omission is VERY DIFFERENT from lying.
Last edited:
Problem. The group of four should lit. Maybe all have been inserted the other way around, or even one inserted wrong can break the whole chain.
TP 1 to ground should go over 7V when the LEDS will eventually operate but no more than 8V. TP 2 to ground should be adjusted to 4V after that. Or TP 1 to TP2 for 3.6V difference, same thing. The PSU side sounds like doing basically well. But restore the four LEDS situation first so we will check the whole behavior better.
I would not trust a TO-92 on 100mA constant for the rest of its service life if I could use stronger in this Cmult. 12mW/C derating factor above 25C I find somewhat precarious for the purpose.
It will do 150mW at 100mA in this application and I would want to see 450mW margin at 50C to feel comfortable (3x). But its just me.
Cruz generally hot rods also. Good steady HFE up to 300mA and the enhanced reliability of a TO-220 2SC5171 should serve him trustfully I would think. Especially when already in his stash.
You are quite right to state Pq limits. But you omitted that from your first warning post
A transistor used for switching with 10mA of base current and 100mA of collector current dissipates approximately
0.1A * 0.1V plus 0.01A * 0.8V, through the two inputs, about 18mW, well below the 200mW limit of a sot23.
The transistor outlast me passing that 100mA.
There are many To92 that will pass 300mA continuously and sot23 can probably manage 200mA easily.
One does not have to go go To220 if one looks at dissipation and selects accordingly.
You said Ic limit and I said "too simplistic"
A transistor used for switching with 10mA of base current and 100mA of collector current dissipates approximately
0.1A * 0.1V plus 0.01A * 0.8V, through the two inputs, about 18mW, well below the 200mW limit of a sot23.
The transistor outlast me passing that 100mA.
There are many To92 that will pass 300mA continuously and sot23 can probably manage 200mA easily.
One does not have to go go To220 if one looks at dissipation and selects accordingly.
