Hi all,
Regarding the Blameless amp that Douglas Self has published, I am wondering if there is any disadvantage to putting some resistance into the collector leg of the first transistor of the emitter follower. This is to limit the current during clipping. I know of the method Douglas normally uses; putting some resistance into the bottom of the VAS and sensing the current there, but thought that a simple resistor is a simpler method. What would the disadvantage of that be?? (if any).
Thoughts ???
Thanks,
Paul.
Regarding the Blameless amp that Douglas Self has published, I am wondering if there is any disadvantage to putting some resistance into the collector leg of the first transistor of the emitter follower. This is to limit the current during clipping. I know of the method Douglas normally uses; putting some resistance into the bottom of the VAS and sensing the current there, but thought that a simple resistor is a simpler method. What would the disadvantage of that be?? (if any).
Thoughts ???
Thanks,
Paul.
Last edited:
I have seen both used. Some form of limiting is essential when using that buffer, as it can dump silly currents into the base of the vas.
I have not seen an analysis of the comparison of alternative limiting methods.
Based on a strong desire to fully control the the currents, even in hard clipping, I prefer the current sense / transistor method.
An Australian magazine published a design using the simple resistor with measured performance that was excellent. So I imagine that the difference won't be massive
I have not seen an analysis of the comparison of alternative limiting methods.
Based on a strong desire to fully control the the currents, even in hard clipping, I prefer the current sense / transistor method.
An Australian magazine published a design using the simple resistor with measured performance that was excellent. So I imagine that the difference won't be massive
Thanks for the reply.
My main concern was whether there will be any distortion consequences from using a collector resistor. My spice simulations say no, but I'd like some other opinions. Unfortunately my knowledge doesn't go that far as to know what possible effect resistance in the collector will do under normal conditions.
Cheers,
Paul.
My main concern was whether there will be any distortion consequences from using a collector resistor. My spice simulations say no, but I'd like some other opinions. Unfortunately my knowledge doesn't go that far as to know what possible effect resistance in the collector will do under normal conditions.
Cheers,
Paul.
What size resistor would be required to protect that transistor given your supply rail and the SOA of that transistor.
Sometimes to correctly limit current, it might take a bigger resistor than most would like to use there.
I might be wrong, to lazy to calculate 🙂
If I remember correctly:
I've used a diode clamp over my darlington VAS, the clamp will protect the small transistor in the VAS but not the large One.
The transistor activated from the emitter resistor of the larger transistor will protect the larger transistor but not the smaller one.
A combination of both will protect both without the need of the collector resistor in the collector of the smaller transistor.
But I think most don't like the effect the clamping diode has on THD, which on the other hand cleans up the effect of driving the amp into clipping.
All above said knowing that I'm a amateur not an expert
Regards
Sometimes to correctly limit current, it might take a bigger resistor than most would like to use there.
I might be wrong, to lazy to calculate 🙂
If I remember correctly:
I've used a diode clamp over my darlington VAS, the clamp will protect the small transistor in the VAS but not the large One.
The transistor activated from the emitter resistor of the larger transistor will protect the larger transistor but not the smaller one.
A combination of both will protect both without the need of the collector resistor in the collector of the smaller transistor.
But I think most don't like the effect the clamping diode has on THD, which on the other hand cleans up the effect of driving the amp into clipping.
All above said knowing that I'm a amateur not an expert
Regards
Last edited:
Its probably not a good idea. Depending on the transistors used, you could well find it introduces hf oscillation. You can even get it to do that in simulation. Its also quite an ill defined way of making a limiter. The resistor would have to be high in value to limit the current and even then there is still current available direct from the front end, more than enough to cause the VAS to draw more than wanted.
So imo not a great method tbh.
So imo not a great method tbh.
Thanks. I think I'll go with the transistor method then.
Can anyone tell me what I should limit the base current of the main VAS transistor to ?? There's no max listed on the data sheet. I have read that max base current can be approximated to 1/3 of the collector current. The max collector current for my VAS driver is 100mA, so would 30-35mA be ok?? I'd like to make the sense resistor as low as posible.
Cheers,
Paul.
Can anyone tell me what I should limit the base current of the main VAS transistor to ?? There's no max listed on the data sheet. I have read that max base current can be approximated to 1/3 of the collector current. The max collector current for my VAS driver is 100mA, so would 30-35mA be ok?? I'd like to make the sense resistor as low as posible.
Cheers,
Paul.
You would never see anything approaching those kind of values. By sensing the VAS emitter current (Dougs method) you automatically limit the base current anyway. Its a non problem. Just use a value that give the max emitter current you desire such as 10 or12 ohm.
Well I've hooked up the transistor, 1K base res from a 12ohm VAS emitter resistor and I'm getting oscillation on negative clipping. Any suggestions ??
Cheers,
Paul.
Cheers,
Paul.
The VAS current increases dramatically at the exact point of clipping.
Its hard to give a definitive reason why your shows this oscillation. Does this oscillation occur without the limiter in place ? Or with no load attached ?
Be very careful testing, and increase the level very slowly to bring the amp just to the point of clipping. It could also be down to construction layout and trace routing... for example if the VAS current (which increases so dramatically) can interfere or modulate other points on the same trace.
Its hard to give a definitive reason why your shows this oscillation. Does this oscillation occur without the limiter in place ? Or with no load attached ?
Be very careful testing, and increase the level very slowly to bring the amp just to the point of clipping. It could also be down to construction layout and trace routing... for example if the VAS current (which increases so dramatically) can interfere or modulate other points on the same trace.
I haven't tried it without any form of limiting in place, only with the simple resistor in the collector where the oscillation didn't show when using a fairly high (24k) resistor. It seems as the available current increases, so does the oscillation. The circuit is fine during normal operation.
I dunno... even the simulation without a limiter can oscillate depending on the "model" used for the VAS. Which is surprising tbh... I suspect in reality it probably would not.
When it clips (at the very point clipping starts) the VAS current jumps up perhaps ten fold, and that could be enough to cause some local feedback via traces and wiring (depending on your layout).
I built the blameless back in the day and on the official boards and never remember any problems like this tbh. I've looked at the original diagrams and this is the limiter.
When it clips (at the very point clipping starts) the VAS current jumps up perhaps ten fold, and that could be enough to cause some local feedback via traces and wiring (depending on your layout).
I built the blameless back in the day and on the official boards and never remember any problems like this tbh. I've looked at the original diagrams and this is the limiter.
Attachments
I've rarely had a "Darlington" Vas that clips cleanly. Maybe it's something I'm doing wrong.
But I've been happy using a diode clamp from collector of 2nd transistor to base of first transistor. The diode I use since I've seen another forum member use it , is BAV21
2KW do you have a way of measuring thd then you could see how diode affects thd .
Other wise use a switch in line with diode and switch it in and out of circuit to see if you can hear a difference.
Regards
But I've been happy using a diode clamp from collector of 2nd transistor to base of first transistor. The diode I use since I've seen another forum member use it , is BAV21
2KW do you have a way of measuring thd then you could see how diode affects thd .
Other wise use a switch in line with diode and switch it in and out of circuit to see if you can hear a difference.
Regards
An approach I like is this MOSFET Vas. No base current and the maximum Vas current is set by R9. The MOSFET acts as a variable resistor in the Q6 current source.
Attachments
Last edited:
If I understand correctly, the diode you are talking about is parallel to miller cap. It does affect THD, because a diode also has capacitance, and its value depends on the its voltage. It will increase distortion of VAS.
It's as you say, now you have to decide if that increase is audible,
And if it's worth the compromise, with benefits comes drawbacks.,
And if it's worth the compromise, with benefits comes drawbacks.,
Apparently the diode capacitance is non linear too. I have tried this, and it did make a difference but the increase in distorion has me against that method.
For now I am going with the resistor in series with the collector of the first transistor. This then also gives me a good takeoff point for my clipping indicator circuit. I currently have no way to detect distortion, but have just bought an Asus Xonar STX card, and with some software will hopefully be able to do some testing....
Cheers,
Paul
For now I am going with the resistor in series with the collector of the first transistor. This then also gives me a good takeoff point for my clipping indicator circuit. I currently have no way to detect distortion, but have just bought an Asus Xonar STX card, and with some software will hopefully be able to do some testing....
Cheers,
Paul
- Status
- Not open for further replies.