Simple Symetrical Amplifier

Status
Not open for further replies.
Jump to Latest
#2,002
a.wayne said:
LC.

Great looking chassis and layout , a couple questions , what is the 4 pin locking connector for ? Any issues with the transformers being so close ?
Click to expand...

4-pin connector conducts 12 V remote trigger input/output signals if connected, so an amp can be turned on by multichannel preamp or similar and at the same time when on sends the on trigger signal to another amp etc. Trigger input is galvanicaly isolated from any amp's potential, even earth connection, to prevent any interference currents inside the system (HT, stereo, etc.). 😎

So far no issue regarding transformers position. Some DIY-ers even stuck toroids without any issue. Because on non-ferrite material of the AL-heatsinks there is none magnetic short connection between transformers. 😉
 
#2,003
Lazy Cat said:
Sometimes fuses saves the day
Click to expand...
Did-you make the screw driver test at 1, 10, 100 watts ? (Big sparkles)
Well, i must agree that, since i managed a big PA system, i'm pretty paranoid.

I must admit that your solution is enough for a domestic use. But it would be better with my way to detect abnormal situations, witch is instant, comparing to your.
 
Last edited:
#2,006
Sheldon said:
If I abandon the Thermaltraks and go with Nico's FET version, are these values OK? Retain 2SA/SC drivers, or use BC550/60? Leave or remove the driver emitter resistors?

I assume the drivers are run at lower current levels than for the BJT version. Am also assuming this version will be easier to stabilize for temp drift.

Sheldon
Click to expand...

550/560 are much OK for input, but avoid for drivers, they may die during bias drifts in test phase many times, Vceo not enough for full swing also. Emitter resistors on drivers did not play a key stabilization role, those were BD139/140 I had experience with though, maybe doing better for your 2SA1360/2SC3423.
Yes they can run <5mA and the amp still be fast. You may need other than 330R to turn the drivers on at the current you want. Don't assume that is too easy to stabilize VS BJT OPT though. That 330R between the output Mosfet gates need be something more elaborate IMHO, there it was a pain to keep bias in mine. Maybe a Vbe multiplier, but have your own test go to see also.
 
#2,012
Salas said:
550/560 are much OK for input, but avoid for drivers, they may die during bias drifts in test phase many times, Vceo not enough for full swing also. Emitter resistors on drivers did not play a key stabilization role, those were BD139/140 I had experience with though, maybe doing better for your 2SA1360/2SC3423.
Click to expand...

Thanks Salas. Yes, bias stability is tricky. If I understand correctly, both the input devices and drivers increase conduction with temperature rise, which amplifies the affect.

Salas said:
Yes they can run <5mA and the amp still be fast. You may need other than 330R to turn the drivers on at the current you want. Don't assume that is too easy to stabilize VS BJT OPT though. That 330R between the output Mosfet gates need be something more elaborate IMHO, there it was a pain to keep bias in mine. Maybe a Vbe multiplier, but have your own test go to see also.
Click to expand...

The Thermaltrak output devices are pretty well compensated by the internal diodes. They seem to be fairly stable to temperature changes, if I correct for the driver current. It's the driver current that is difficult to stabilize, so I guess that you are probably correct that a VBE multiplier would be needed to stabilize the FET version. In that case the output devices are self stabilizing, but the VBE multiplier would compensate for the driver drift. However, that means that the current through the drivers is still allowed to vary a lot - which could affect distortion. So how do we get the drivers to quickly stabilize to a desired current? It seems like that is the issue. Is LC's latest scheme the only way?
 
#2,013
I used a JFET CCS under the upper driver and fixed their bias, some argued that goes against full symmetry, others argued its gonna effect it only near clip. To me the most important aspect is not to force too much base current to the upper driver, its going to suck it up from the respective input BJTs and be exhausting the drive ability non symmetrically with swing then.
 
#2,014
As I mentioned before, I considered the NTC in the early days of designing the TM5 to 'fix' the bias current stability. The place I found worked best was in place of the resistor that feeds current from the zener diodes to the feedback node. I found suitable thermistors at digikey with the right coefficient to hold it stable. But I didn't use them in the end, I don't think they are necessary - and they are expensive. Perhaps for the big higher power amps it becomes necessary again.

LC - I like the way your build is coming along, nice mechanical layout. Did you orient the power trafo's so that their magnetic flux leakage cancels out ? (I saw that done in a Goldmund amp I think).
 
#2,015
Salas said:
I used a JFET CCS under the upper driver and fixed their bias, some argued that goes against full symmetry, others argued its gonna effect it only near clip. To me the most important aspect is not to force too much base current to the upper driver, its going to suck it up from the respective input BJTs and be exhausting the drive ability non symmetrically with swing then.
Click to expand...

Will this work with BJT outputs?

Sheldon
 
#2,020
Status
Not open for further replies.
Top Bottom