Hi,
It has been a few years since i was on here last, Lack of time and moving around had taken its toll on my tinkering. I am hopeing that some of the old faces are still around, some of them might even remember my original thread back in 2005 about this amp?
Anyway, I was clearing out my dads house after he moved into sheltered housing and i came accross my old hybrid amp, It had taken a beating in storage and after swapping out the unfotunatly smashed ECC88's only one channel was working, i traced the fault back to a bad connection in the power supply and it sprang to life.
I decided i would make a schematic for it and try and get some more tikering done, i never did finish the amp, i remember i tried CCS instead of the botstraps, and i tried various different output transistors, etc. This circuit diagram is exactly how the amp was when i left it back in 2006 (schematic was reverse engineered from the amp). The build quality of the amp was pretty apauling and i am going to have to scrap it and start again.
So, Can anyone make any suggestions for improvements to the design, any suggestions on good replacements for the BD139 / BD140? I remember trying MJE340 / MJE350 but i am guessing i liked the BD's better or burnt out the MJE's because it has BD's in it at the moment.
It has been a few years since i was on here last, Lack of time and moving around had taken its toll on my tinkering. I am hopeing that some of the old faces are still around, some of them might even remember my original thread back in 2005 about this amp?
Anyway, I was clearing out my dads house after he moved into sheltered housing and i came accross my old hybrid amp, It had taken a beating in storage and after swapping out the unfotunatly smashed ECC88's only one channel was working, i traced the fault back to a bad connection in the power supply and it sprang to life.
I decided i would make a schematic for it and try and get some more tikering done, i never did finish the amp, i remember i tried CCS instead of the botstraps, and i tried various different output transistors, etc. This circuit diagram is exactly how the amp was when i left it back in 2006 (schematic was reverse engineered from the amp). The build quality of the amp was pretty apauling and i am going to have to scrap it and start again.
So, Can anyone make any suggestions for improvements to the design, any suggestions on good replacements for the BD139 / BD140? I remember trying MJE340 / MJE350 but i am guessing i liked the BD's better or burnt out the MJE's because it has BD's in it at the moment.
Attachments
Last edited:
Hi,
use triple darlington output, avoid the bootstrap from the ss stage input. This is what I found in the last 2 years since I started to use this topology.
Sajti
use triple darlington output, avoid the bootstrap from the ss stage input. This is what I found in the last 2 years since I started to use this topology.
Sajti
Interesting idea. Will the Triple Darlington configuration provide a high enough input impedance to keep the tube stage happy without the bootstrap though?
Do you mean Drop the BD139/140 and replace with a darlington like tip120/125 to create a triple tier on the output or did you mean to do it by adding another transistor in?
I have a whole load of spare output transistors, i was toying with the idea of trying a class AC output stage but i think i might keep that idea for a new project at some point in the future.
Do you mean Drop the BD139/140 and replace with a darlington like tip120/125 to create a triple tier on the output or did you mean to do it by adding another transistor in?
I have a whole load of spare output transistors, i was toying with the idea of trying a class AC output stage but i think i might keep that idea for a new project at some point in the future.
Not sure I'd go to triple EF, a bit tetchy and prone to oscillation, but I'd consider an npn EF at the cathode of the upper ECC88 to strengthen drive, and increasing current through the double bootstrapped bias network to around 7mA to feed the drivers.
I'd also move to 2SC4793/2SA1837 drivers; higher bandwidth, more current, and higher beta as well.
Apart from that, I think it's an excellent design and it should sound very good.
Cheers,
Hugh
I'd also move to 2SC4793/2SA1837 drivers; higher bandwidth, more current, and higher beta as well.
Apart from that, I think it's an excellent design and it should sound very good.
Cheers,
Hugh
Hi Hugh,
That is more the sort of suggestions i was looking for. I will have to try them (in the simulator) when i get home.
I had a little play increasing the current through the bootstrapped bias network last night, I got it to ~3mA before i gave up and went to bed, Initial results in the simulator looked pretty good though. Will decreasing the resistors too much not have a negative affect on the input impedance or will
the EF on the top ECC88's cathode mean that it isnt really a problem anymore?
I guess i will find out for myself later on!
Cheers,
Owen
My amplifier use NJW0*** output devices, 2SC5171/2SA1930 drivers, and 2SC3423/2SA1360 predrivers. The predrivers are running with 2.5mA bias, which means about 20uA base current, so I have choosed the current of the input bias resistors 400uA. This means 100kohm resistors, and about 50kohms input impedance. This is easy to drive with tube.
Sajti
Last edited:
I had no problem with that. I use base stoppers 100 ohms for predriver, 22ohm for the driver, and 10ohms for the output devices. I also use 1uH series with the load.
Of use some stronger tubes as cathode follower. 6H6 would be nice. Or some MOSFET is suitable as well.
Agree.
Sajti
Last edited:
I ran the circuit in LTspice and it seems that there is not enough voltage gain in the tube stage to drive the transistor stage to full power.
I will have a look what i have in my parts box, might have some different tubes but most of what i have are ecc88 and equivilents these days. going to try changing the drivers out first as soon as i manage to get my hands on some of those that were suggested, only place in UK i can find them is ebay and i dont trust they will be genuine if i buy them on ebay.
When you ran the simulation did you set up the bias and dc offset? bias should be about 50ma each side through the outputs from what i remember, not sure if it is necessary but i think its a possibility that different tube / transistor models could mess up the settings.
Did you happen to notice with near zero (under 1mv) dc offset you still get offset of ac wave on output? The actual amp does not suffer from this but any suggestions why it might in LTspice?
I am aware you cant drive the output stage to full power, it was designed and built around stuff i had lying around at the time. I never really saw it as a problem.
- Status
- Not open for further replies.