It looks possible: Terranigma has asked the same question, and I gave an example circuit here (tested in sim only):
http://www.diyaudio.com/forums/solid-state/189599-my-little-cheap-circlophone-115.html#post3354407
I'm building a darlington circlophone following Elvee's example circuit mentioned above. I think it is going to be first darlington build. By considering your requirements, MJ2500 is out of spec. It is PNP and Vceo is 60V according to Motorola datasheet. Our circuit implies NPN power darlington has at least 2*Vs Vceo. You have to look for NPN type output devices at least 100V Vceo according to your 45V rails.
It would be possible to use a P-darlington as a drop-in replacement for each CFP.
The B-E resistor just needs to be raised to 120ohm, like here:
http://www.diyaudio.com/forums/solid-state/189599-my-little-cheap-circlophone-2.html#post2585337
The B-E resistor just needs to be raised to 120ohm, like here:
http://www.diyaudio.com/forums/solid-state/189599-my-little-cheap-circlophone-2.html#post2585337
Whoops my bad, yep I have corrected that I have inter-mixed pnp to npn while describing the circlophone to him, actually he was puzzled when he mentioned to me the MJ2500 he had...
Nice to know that you will be creating a Darlington version.
TY!
Hello! I am currently working on this particular model, but am having a bit of an issue. The resistor in the middle of the board - the one surrounded by the coil - is having its value blocked. I cannot see what it says. I think it is R17 on the original schematic, which is rated at 10K. Is this correct? On the board, it kinda looks like it says 10/3W, as in, just 10R. This is making me confused.
Thanks everybody, for all of this. Very exciting thing, this Circlophone.
Hi again! I'm still working on Piersma's CFP circuit. But I have some more questions.
1. What is the value of this resistor? On Piersma's picture, I can see that there's a resistor here, rather than a jumper, which I originally assumed the 000 ohm designation meant. I cannot see the value of it, though, as the picture is too low res.
2. I assume these unmarked diodes are supposed to be 1N4148, but I would like to make sure, just in case.
3. I do not have 11K5 resistors. Would 12K be okay to use here?
4. Is this coil critical? It says 22 turns, but what else do I need to keep in mind? Wire thickness? Coil diameter?
Sorry for all this. I'm learning! Thanks for reading.
1. What is the value of this resistor? On Piersma's picture, I can see that there's a resistor here, rather than a jumper, which I originally assumed the 000 ohm designation meant. I cannot see the value of it, though, as the picture is too low res.
2. I assume these unmarked diodes are supposed to be 1N4148, but I would like to make sure, just in case.
3. I do not have 11K5 resistors. Would 12K be okay to use here?
4. Is this coil critical? It says 22 turns, but what else do I need to keep in mind? Wire thickness? Coil diameter?
Sorry for all this. I'm learning! Thanks for reading.
I somehow missed this when ordering parts... And now I cannot for the life of me find a 2SC2911 anywhere. What other transistors could I use here? Would BD139 work? I'm going to be using a +/-30V supply.
Thanks again.
I somehow missed this when ordering parts... And now I cannot for the life of me find a 2SC2911 anywhere. What other transistors could I use here? Would BD139 work? I'm going to be using a +/-30V supply.
Thanks again.
I tested BD's for darlington output version and they seem OK (32V+32V for me). But mine were good ones from Philips and Harris. BD devices varies according to their manufacturers. I suggest you to use 2sc5171 as mentioned on first post and tested by Elvee. But take attention to their leg arrangement. E-C-B for 2sc2911 and B-C-E for 2sc5171 which means you can substitute 2sc5171 just flipping without bending their legs.
Last edited:
Piersma,
Could you do us the gigantic favor of re-scaling your clean treble version of Circlophone CFP to use 35+35VDC rails (range 32+32vdc to 38+38vdc) and with a gain of at least 42X (range 42X to 50X)? This will use the capacity of the CFP design, self-amplify dynamics, and result in "Circlophone Turbo" amplifier?
The greater differentiation sounds like more power and more headroom, no matter if it is LM1875/4702 turbo or Honey Badger turbo--more peak power is more dynamic. Those are miller types. Shunt comp types *could be* even more lively. Circlophone CFP edition was born for this use, but I see that the feature for which it was made, has not been used.
Setting the CFP to low gain is like gluing a tennis ball underneath the accelerator pedal of a sports car--the capacity is actually there but the thrill is blocked. Can you free it?
This may need both resistors of the feedback voltage divider readjusted for value and probably needs the compensation tweaked. I'm not qualified to do that for the CFP edition, but you can do it nicely. So, I'm asking for your help.
Could you do us the gigantic favor of re-scaling your clean treble version of Circlophone CFP to use 35+35VDC rails (range 32+32vdc to 38+38vdc) and with a gain of at least 42X (range 42X to 50X)? This will use the capacity of the CFP design, self-amplify dynamics, and result in "Circlophone Turbo" amplifier?
The greater differentiation sounds like more power and more headroom, no matter if it is LM1875/4702 turbo or Honey Badger turbo--more peak power is more dynamic. Those are miller types. Shunt comp types *could be* even more lively. Circlophone CFP edition was born for this use, but I see that the feature for which it was made, has not been used.
Setting the CFP to low gain is like gluing a tennis ball underneath the accelerator pedal of a sports car--the capacity is actually there but the thrill is blocked. Can you free it?
This may need both resistors of the feedback voltage divider readjusted for value and probably needs the compensation tweaked. I'm not qualified to do that for the CFP edition, but you can do it nicely. So, I'm asking for your help.
Last edited:
That's cool.
An example of advanced miller comp "turbo" amp, is Honey Badger, 42X. Sounds like a lot of headroom. It will do that sound even if severely undervolted. That sound of big headroom is actually "sufficient power" in combination with high gain.
To run high gain needs an amp with a lot of open loop gain to spare, and thus Circlophone CFP was made for this use. As entertaining as a miller might be, I'd love to hear a shunt comp turbo. Circlophone Turbo! Much lively dynamics for maximizing your music collection. That's what its for. High gain is just a cost/means of doing it effectively. The CFP is like a sports car, but we didn't use the main feature yet. Amazement awaits. . .
Last edited:
Hi guys,
I finally built my Circlophone. As mentioned earlier, I used Piersma's CFP PCB design.
Now, problem.
First, I am using a +/-15V supply (zeners and resistor are correct for the application).
No sound comes out. I've been probing around with a scope, but I am not an expert in this. Not sure how to properly troubleshoot doing this. With a sine wave to the input, I am able to see (an amplified) signal on the emitters of Q10 and Q11 on the pcb layout. Those are the 2SC2911 transistors. I did not have those, but instead used 2N5551 (hFE around 200). The voltage at the bases of these are seemingly -15V.
Note: The signal I see is severely clipped, pk-pk is 5,88V. The signal generator's output is 50 ohm and looks fine in the beginning of the amplifier (before it has been amplified).
There's no signal on the collectors of these two transistors (phase inverters)?). I'm thinking there should be. I am not sure if this is the issue.
Also, there seems to be no voltage across R4, R5, R28 or R29. No idle current?
Any ideas on what might be wrong, or how to troubleshoot this? Thanks a lot for your time.
I finally built my Circlophone. As mentioned earlier, I used Piersma's CFP PCB design.
Now, problem.
First, I am using a +/-15V supply (zeners and resistor are correct for the application).
No sound comes out. I've been probing around with a scope, but I am not an expert in this. Not sure how to properly troubleshoot doing this. With a sine wave to the input, I am able to see (an amplified) signal on the emitters of Q10 and Q11 on the pcb layout. Those are the 2SC2911 transistors. I did not have those, but instead used 2N5551 (hFE around 200). The voltage at the bases of these are seemingly -15V.
Note: The signal I see is severely clipped, pk-pk is 5,88V. The signal generator's output is 50 ohm and looks fine in the beginning of the amplifier (before it has been amplified).
There's no signal on the collectors of these two transistors (phase inverters)?). I'm thinking there should be. I am not sure if this is the issue.
Also, there seems to be no voltage across R4, R5, R28 or R29. No idle current?
Any ideas on what might be wrong, or how to troubleshoot this? Thanks a lot for your time.
Last edited:
Sorry for spamming up the place before giving up on my investigation. Turned out that the transistors really were the issue. I moved the legs of the 1893s to their intended destinations, and everything got sorted. I even built another one that worked right off the bat.
I made some mistakes along the way, but overall, this was a fun and easy project. Thanks again to Elvee for this thing, and Piersma for the layout and everyone else involved in this Circlophone.
The wire I used for the coil is rather thin. I intend to replace it later, but with what? And how long? On Piersma's picture there are only 12 turns, but I have 22 as seen on the PCB layout. Does it even matter?
Some pics of the (mostly) finished product (seems oddly scaled by Diyaudio. Click for full sizes) :
I made some mistakes along the way, but overall, this was a fun and easy project. Thanks again to Elvee for this thing, and Piersma for the layout and everyone else involved in this Circlophone.
The wire I used for the coil is rather thin. I intend to replace it later, but with what? And how long? On Piersma's picture there are only 12 turns, but I have 22 as seen on the PCB layout. Does it even matter?
Some pics of the (mostly) finished product (seems oddly scaled by Diyaudio. Click for full sizes) :
Very Nice build!
Those white color pcb standoff looks like plastic material..they should be made of metal/brass type, the heatsink could easily melt them (unless they are made to withstand heat).
The coil is not so critical but for heavy duty purpose, use a larger wire you could just wind it over the entire length of the resistor..doesnt matter if its 12 or 22 turns.
TY!
Those white color pcb standoff looks like plastic material..they should be made of metal/brass type, the heatsink could easily melt them
(unless they are made to withstand heat).The coil is not so critical but for heavy duty purpose, use a larger wire you could just wind it over the entire length of the resistor..doesnt matter if its 12 or 22 turns.
TY!