Thats the arrangement I am currently using in all my amps. I use CCS' which seem to work well. I have a 1meg resistor from the caps tied node to keep them polarized. What value did you find worked best for you Allen ?
I like this arrangement because it is self adjusting. I appreciate that its going to be slightly sound compromised but the thought of fiddling with bias adjustment every month or so leaves me cold. I understand that the 300dpa need manual adjustment.
The problem I am currently having is at power up the output tube is zero or positively biased and asking to pass huge currents, but the dinky little MJE350's can't cope with the current slam - hence pop. Its a particular facet of the DC coupling I am using. I need to find a good transistor with 600V capability and about 4amps current capability. The nasty thing is because the CCS are tied together through the big differential caps, if one CCS goes down then the other instantly follows.
Shoog
Shoog,
I have never had any luck with using bipolar trs under hi voltage condidtions. I always now use MOSFETs as , in my experience, they are VASTLY more reliable. I've had them so hot even with 350v across them the black burnt off the heatsinks (!!!) but they keep on running just fine - can't do that with bipolars.
I find the optimum value for the R to gnd from the center tap of the cathode C's by distortion measurements, start around 100ohms and go up or down from there.
Regards, Allen
I have never had any luck with using bipolar trs under hi voltage condidtions. I always now use MOSFETs as , in my experience, they are VASTLY more reliable. I've had them so hot even with 350v across them the black burnt off the heatsinks (!!!) but they keep on running just fine - can't do that with bipolars.
I find the optimum value for the R to gnd from the center tap of the cathode C's by distortion measurements, start around 100ohms and go up or down from there.
Regards, Allen
Hey Shoog,
Why don't you try Gary Pimms CCS - reportedly excellent sounding. It can take the voltage but I'm not so sure about the 6A current? He has Eagle files on his site for PCB ordering.
I'm going to order some DN2540 depletion mode Mosfets from Mouser in the next couple of days - let me know if you want in! Could also pick up some good Fets too.
I've just built his CCS using LND150s & John Swenson's 250 & 130V voltage regulators based on this CCS for the BDT preamp. They're a pain to do P2P so pcbs would be of interest to me too!
Why don't you try Gary Pimms CCS - reportedly excellent sounding. It can take the voltage but I'm not so sure about the 6A current? He has Eagle files on his site for PCB ordering.
I'm going to order some DN2540 depletion mode Mosfets from Mouser in the next couple of days - let me know if you want in! Could also pick up some good Fets too.
I've just built his CCS using LND150s & John Swenson's 250 & 130V voltage regulators based on this CCS for the BDT preamp. They're a pain to do P2P so pcbs would be of interest to me too!
Albert,
ZVN0545 in the design dissipate only about 50mW or less so heatsink is not required. I had a huge vinyl session with these amps over the weekend. Showed that these are the best amps I've had on the system so far BUT can't rest on my laurels - there is still plenty of room for improvement.
Interestingly enough - when listening to CDs (that digital stuff) I prefer the "Jessica" (6V6G with 6SL7) version of the amp. Its not technically as good but it does a nice job of covering up the digital artifacts and so sounds better with CDs.
Cheers,
Ian
ZVN0545 in the design dissipate only about 50mW or less so heatsink is not required. I had a huge vinyl session with these amps over the weekend. Showed that these are the best amps I've had on the system so far BUT can't rest on my laurels - there is still plenty of room for improvement.
Interestingly enough - when listening to CDs (that digital stuff) I prefer the "Jessica" (6V6G with 6SL7) version of the amp. Its not technically as good but it does a nice job of covering up the digital artifacts and so sounds better with CDs.
Cheers,
Ian
I rebuilt my CCS last night and plugged her in. Two power ups so far and she's OK. Not really as robust as I would like but baring a fault situation elsewhere in the amp they should last.
Current balance isn't quite as good as I am used to, and this may be effecting the bass which sounded a bit ropey last night.
The only problem with the Gary Pimm CCS is that they are very complex compared to all the others, and the level of performance they offer just isn't warrented in a cathode duty with bypass caps.
Shoog
Current balance isn't quite as good as I am used to, and this may be effecting the bass which sounded a bit ropey last night.
The only problem with the Gary Pimm CCS is that they are very complex compared to all the others, and the level of performance they offer just isn't warrented in a cathode duty with bypass caps.
Shoog
Shoog,
Glad to hear it's working - so you did something different? Are your power up issues as a result of the quasi differential O/P stage or elsewhere in the topology? I want to try this O/P configuration on my Baby Huey & just need to be aware of potential probs
I agree about GP's CCS being complex but if you consider the problems you've had with this CCS ........ maybe not such a big build?
Maybe even consider cascoded DN2540 if this CCS fails!
Glad to hear it's working - so you did something different? Are your power up issues as a result of the quasi differential O/P stage or elsewhere in the topology? I want to try this O/P configuration on my Baby Huey & just need to be aware of potential probs
I agree about GP's CCS being complex but if you consider the problems you've had with this CCS ........ maybe not such a big build?
Maybe even consider cascoded DN2540 if this CCS fails!
Ian,
Thanks for the infos.
I've ordered the ZVN0545A and they're on the way to me. I'll go for the fixed bias in this monment so to compare the baby huey.
I also have some 6V6 RCA and some 6SL7 RCA on my collection.
I certainly will give a it a try in the later date.
About the sound of the fixed bias EL84, Can we play a bit on the feed back resisitor? I mean to lower the value to have smoother / warmer sound.
Albert
Thanks for the infos.
I've ordered the ZVN0545A and they're on the way to me. I'll go for the fixed bias in this monment so to compare the baby huey.
I also have some 6V6 RCA and some 6SL7 RCA on my collection.
I certainly will give a it a try in the later date.
About the sound of the fixed bias EL84, Can we play a bit on the feed back resisitor? I mean to lower the value to have smoother / warmer sound.
Albert
Fixed bias baby huey
Ian,
I have my fixed bias baby huey up for testing with the following numbers:
B+ 311v
plate 1 & 2 303v
G2 1 & 2 306v
Bias both 0.40v
+51v for ZVN0545A
B- 40v
On 12ax7 plate 1 61.5v
plate 2 -3.2v ( this is something not right )
Both green and blue LEDs don't come on. I did pull those out and check. They are ok.
Thanks
Ian,
I have my fixed bias baby huey up for testing with the following numbers:
B+ 311v
plate 1 & 2 303v
G2 1 & 2 306v
Bias both 0.40v
+51v for ZVN0545A
B- 40v
On 12ax7 plate 1 61.5v
plate 2 -3.2v ( this is something not right )
Both green and blue LEDs don't come on. I did pull those out and check. They are ok.
Thanks
Attachments
Albertli,
The only way you could have a negative voltage on the anode of that 12AX7 is a wiring error or shorted coupling cap across to the ZVN0545.
Also the 2 LEDS MUST be ON for the CCS to work. The 0V to resistor, through 2 LEDs to -ve rail is an easy path to check. If you have CCS transistors of wrong type (PNP instead of NPN) or wired wrongly they could have a forward biased junction shorting the current around the LEDs in which case you would measure 0.7 Volts across the LEDs instead of about 1.8 to 2.0 V for the green and 2.3V for the blue.
I would be doing a full check of the front end wiring, looking for something not right.
Cheers,
Ian
The only way you could have a negative voltage on the anode of that 12AX7 is a wiring error or shorted coupling cap across to the ZVN0545.
Also the 2 LEDS MUST be ON for the CCS to work. The 0V to resistor, through 2 LEDs to -ve rail is an easy path to check. If you have CCS transistors of wrong type (PNP instead of NPN) or wired wrongly they could have a forward biased junction shorting the current around the LEDs in which case you would measure 0.7 Volts across the LEDs instead of about 1.8 to 2.0 V for the green and 2.3V for the blue.
I would be doing a full check of the front end wiring, looking for something not right.
Cheers,
Ian
Ian,
Thanks for the quick reply. I certainly will check the front end wiring tomorrow.
Being a complete newbie, I did follow closely by using the exact parts ( NPN BC546B ) as of your schematic. Now the problem is I don't know how to check if they are ok.
The coupling caps are rated at 100v, so I thought it should be fine for this application.
Regards
Albert
Thanks for the quick reply. I certainly will check the front end wiring tomorrow.
Being a complete newbie, I did follow closely by using the exact parts ( NPN BC546B ) as of your schematic. Now the problem is I don't know how to check if they are ok.
The coupling caps are rated at 100v, so I thought it should be fine for this application.
Regards
Albert
Albert,
100V rated coupling capacitors are NO GOOD.
At power on, before the tubes start conducting you may have as much as 350 or even 370 Volts DC on the anodes of the 12AX7. The other side of the caps have negative bias voltage on them. There is probably a total of around 400V across the caps. Even when the amp is operating correctly (when the 12AX7 anode volts will drop down) you will have about 200V across the coupling caps. They are almost certainly Cactus or whatever the expression to describe "gone to god" is in your part of the world. You will need to replace them. I would use 630V rated caps. Do this before chasing anything else.
Note: A "Cactus" Cap would cetainly explain that -3.2 volts you saw on one 12AX7 anode.
Cheers,
Ian
100V rated coupling capacitors are NO GOOD.
At power on, before the tubes start conducting you may have as much as 350 or even 370 Volts DC on the anodes of the 12AX7. The other side of the caps have negative bias voltage on them. There is probably a total of around 400V across the caps. Even when the amp is operating correctly (when the 12AX7 anode volts will drop down) you will have about 200V across the coupling caps. They are almost certainly Cactus or whatever the expression to describe "gone to god" is in your part of the world. You will need to replace them. I would use 630V rated caps. Do this before chasing anything else.
Note: A "Cactus" Cap would cetainly explain that -3.2 volts you saw on one 12AX7 anode.
Cheers,
Ian
Ian,
I replaced the coupling caps to .015 630v and have the measurement below:
1). both cathode of 12ax7 around -4v
2). across green led -18.2v
3). across blue led -7.8v
I have -7.5v between the coupling cap and 220R. -10.8 after the 12v zener.
Albert
BTW the supplies remain the same +51v and -40v.
I replaced the coupling caps to .015 630v and have the measurement below:
1). both cathode of 12ax7 around -4v
2). across green led -18.2v
3). across blue led -7.8v
I have -7.5v between the coupling cap and 220R. -10.8 after the 12v zener.
Albert
BTW the supplies remain the same +51v and -40v.
Albert,
It looks like you've fixed the coupling cap problems, those voltages you quote for "between the coupling cap and the 220R" and the "after the zener" (that is on the source of the ZVN0545A) look good.
You still have a Current Source problem however.
Working from your -40V supply:
1) You should have approx. 1mA through the CCS transistors PLUS about 2.5 to 3 mA through the LED biasing chain. That is approx 3.5 to 4mA total.
That means an approx. 8 volt drop across the 2K2 to give about -32V across the 100uF capacitor.
2) There should be 1.8 to 2.0 volt drop across the 1K8
3) There should be 2.5V drop across the blue led and 2.0V across the green led leaving 26 or 27 volts across the 10K resistor to ground (this resistor sets the led bias chain current).
Junction of the Emitter of the "bottom" transistor and the 1K8 should be at about -30V
Emitter of the "top" transistor should be at about -27V.
Pull the 12AX7 from its socket and work on the CCS till it looks right.
The LEDS should both be ON.
Check for transistors and LEDs incorrectly wired or blown up.
Fault finding tricks:
1) Short across the green LED and make sure the BLUE LED lights. If not, its in backward or the bottom transistor is incorrectly wired or blown.
Once that is going:
Remove the short across the green LED - check that Green LED is ON, if not its in backward or the top transistor is incorrectly wired or blown.
Good luck,
Ian
It looks like you've fixed the coupling cap problems, those voltages you quote for "between the coupling cap and the 220R" and the "after the zener" (that is on the source of the ZVN0545A) look good.
You still have a Current Source problem however.
Working from your -40V supply:
1) You should have approx. 1mA through the CCS transistors PLUS about 2.5 to 3 mA through the LED biasing chain. That is approx 3.5 to 4mA total.
That means an approx. 8 volt drop across the 2K2 to give about -32V across the 100uF capacitor.
2) There should be 1.8 to 2.0 volt drop across the 1K8
3) There should be 2.5V drop across the blue led and 2.0V across the green led leaving 26 or 27 volts across the 10K resistor to ground (this resistor sets the led bias chain current).
Junction of the Emitter of the "bottom" transistor and the 1K8 should be at about -30V
Emitter of the "top" transistor should be at about -27V.
Pull the 12AX7 from its socket and work on the CCS till it looks right.
The LEDS should both be ON.
Check for transistors and LEDs incorrectly wired or blown up.
Fault finding tricks:
1) Short across the green LED and make sure the BLUE LED lights. If not, its in backward or the bottom transistor is incorrectly wired or blown.
Once that is going:
Remove the short across the green LED - check that Green LED is ON, if not its in backward or the top transistor is incorrectly wired or blown.
Good luck,
Ian