PERSPEX QUASAR ver 2.0 prototype build

Dedicated to those large boxes at one end of the room
User avatar
Nick
Site Admin
Posts: 15705
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#166 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Nick »

For that matter, an alps in the feedback loop will have more effect than a cap on the input IMHO.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
Ray P
No idea why I do this anymore
Posts: 6294
Joined: Thu Nov 22, 2007 5:18 pm
Location: Somerset

#167 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Ray P »

Nick wrote: Wed Apr 07, 2021 8:51 am For that matter, an alps in the feedback loop will have more effect than a cap on the input IMHO.
Nick, not an issue for my project - this schematic is the output from Wolfgang's prototype build and I won't have a potentiometer in the feedback loop on my build - I've got some multi-turn potentiometers to find the sweet spot, after which I'll replace with simple good quality resistors.
Sorry, I couldn't resist!
User avatar
Ray P
No idea why I do this anymore
Posts: 6294
Joined: Thu Nov 22, 2007 5:18 pm
Location: Somerset

#168 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Ray P »

Nick wrote: Wed Apr 07, 2021 8:48 am I still think adding a low pass filter inside a feedback loop is going to have a worst effect than a cap in the input.
Isn't there already a hi-pass filter inside the feedback loop Nick, or do you mean that changing the operating point of that filter will alter the optimal value of the required feedback resistor?
Sorry, I couldn't resist!
User avatar
Ray P
No idea why I do this anymore
Posts: 6294
Joined: Thu Nov 22, 2007 5:18 pm
Location: Somerset

#169 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Ray P »

Nick wrote: Wed Apr 07, 2021 8:50 am BTW, what’s the point of the cathode being stood up a bit on the second triode? Is it just reducing gain a bit (for some reason).
I'm not clever enough with this stuff to be able to answer that Nick, my ability is not much beyond 'painting by numbers' - the input/driver stage is lifted from one of John Broskie's TCJ articles, Aikido Cascade IIRC - I'll see if I can find the reference.
Sorry, I couldn't resist!
User avatar
Nick
Site Admin
Posts: 15705
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#170 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Nick »

My point is that the existing hi pass is assumed to be outside the area of signal that matters. Reducing the gain inside the feedback look will have the effect of reducing the cancelation of distortion so may have unwanted effects outside the wanted area.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
Ray P
No idea why I do this anymore
Posts: 6294
Joined: Thu Nov 22, 2007 5:18 pm
Location: Somerset

#171 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Ray P »

Here's the reference to the Broskie input/driver stage - its in the 'Boosted 300B Single-ended Power Amplifier' article (Aikido Cascade ver 3).

https://tubecad.com/2014/03/blog0283.htm
Sorry, I couldn't resist!
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#172 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by IslandPink »

Nick wrote: Wed Apr 07, 2021 8:48 am I still think adding a low pass filter inside a feedback loop is going to have a worst effect than a cap in the input.
Ah sorry, forgot this had feedback. In that case you are right.
"Once you find out ... the Circumstances ; then you can go out"
User avatar
Ray P
No idea why I do this anymore
Posts: 6294
Joined: Thu Nov 22, 2007 5:18 pm
Location: Somerset

#173 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Ray P »

Nick wrote: Wed Apr 07, 2021 9:29 am My point is that the existing hi pass is assumed to be outside the area of signal that matters. Reducing the gain inside the feedback look will have the effect of reducing the cancelation of distortion so may have unwanted effects outside the wanted area.
OK, thanks Nick, on the input it is then.

Image
Sorry, I couldn't resist!
User avatar
Nick
Site Admin
Posts: 15705
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#174 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Nick »

Ray P wrote: Wed Apr 07, 2021 9:43 am Here's the reference to the Broskie input/driver stage - its in the 'Boosted 300B Single-ended Power Amplifier' article (Aikido Cascade ver 3).

https://tubecad.com/2014/03/blog0283.htm
Ok, that doesn't really describe why either.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#175 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by JamesD »

This is an aikido stage and JB has arranged so that the second stage cathode resistor split introduces anti p0hase power supply noise compared to the first stage so as to cancel that noise at the second stage output - its psu noise reduction for the first stage psu noise given that it is running at relatively high gain with low psrr.

He does explain it if you read back far enough but as often with his articles you have to follow through several examples and sometimes previous articles to get the full explanation and for this to work properly the valves have to be working with the parameters assumed in his aikido calculation...
User avatar
Nick
Site Admin
Posts: 15705
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#176 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Nick »

Oh yes, thanks.
the valves have to be working with the parameters assumed in his aikido calculation...
And also to be linear, which of course they are not which is where most of the null things out ideas fall down.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#177 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by JamesD »

Indeed - although to be fair to JB - the Aikido designs will still improve on the situation with regard to psu noise... just not as much as simulation/calculation. Its the distortion cancellation stages that really are variable - some make real life situations worse! That's why I like JBs original aikido design as it used the transfer curve both the right way up and inverted so it did reduce distortion if you used dual triode valves in vertical pairs - assuming better than 20% matching between the sections... and with 10% matching it is quite good...
User avatar
Nick
Site Admin
Posts: 15705
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#178 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by Nick »

JamesD wrote: Wed Apr 07, 2021 5:13 pm Indeed - although to be fair to JB - the Aikido designs will still improve on the situation with regard to psu noise... just not as much as simulation/calculation. Its the distortion cancellation stages that really are variable - some make real life situations worse! That's why I like JBs original aikido design as it used the transfer curve both the right way up and inverted so it did reduce distortion if you used dual triode valves in vertical pairs - assuming better than 20% matching between the sections... and with 10% matching it is quite good...
Yep, it worked better with the second stage not producing gain, that gave closer matching, but that not the case here.

Even so, the cancellation will have a tendency to reduce lower order distortion more than it does higher order ones. Likewise low frequency psrr is improved at the expense of generating harmonics of the power supply noise.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#179 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by JamesD »

Even so, the cancellation will have a tendency to reduce lower order distortion more than it does higher order ones. Likewise low frequency psrr is improved at the expense of generating harmonics of the power supply noise.
That is a really astute observation and accounts for the people who observe that the Aikido can sound a little harsh compared to a non Aikido version...

Neat!

And doing this within a feedback loop is going to exacerbate the situation regarding higher harmonics too... interesting...
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#180 Re: PERSPEX QUASAR ver 2.0 prototype build

Post by IslandPink »

Aha .... that was me, for one. It was very revealing of valve brands/eras as i remember, eg. with the older/newer 6N6P's.
"Once you find out ... the Circumstances ; then you can go out"
Post Reply