Well, passive radar, while interesting, is not what everybody wants to talk about, it seems.
While the A21 has done well in the audio marketplace, it is not my personal design. It is a mixture of what Taiwan thinks is the best way to design a power amp and my general input. It is a 'minimalist' design in many ways, compared to the earlier HCA2200, 2500 (what I am using at the moment), 3500, and finally the JC-1. The delivered price of this unit is about $500, or about 1/4 what a JC-1 costs. The extra cost/Watt of the JC-1 is because of 'refinement' that we put into the schematic and parts quality. Unfortunately, it makes it really expensive to produce.
I wish that I could make a JC-1/2, that would be 1/2 the power of the JC-1 at 1/2 cost, with the 'refinements' of the JC-1 included. That would be the ideal amp for my needs.
I wish that I could make a JC-1/2, that would be 1/2 the power of the JC-1 at 1/2 cost, with the 'refinements' of the JC-1 included. That would be the ideal amp for my needs.
Perhaps it would be useful to take the A21 as is, and 'improve' it into the JC-1/2. This is not completely practical with the existing unit, but it would be a good design exercise. Perhaps a MK-2 version, some time in future.
Oh, of course, it is good value for money, in fact, for many 'objective scientists' it is good enough for just about everybody.
It might be interesting to note that the HCA-3500 is a FAILED design, with a poor subjective rating, but the A-21, while obviously more 'compromised' is a SUCCESSFUL design. Why? That is a fundamental question that we can discuss. However, UNTIL the HCA-3500 was modified into the CTC Bar-b-que power amp, we did not use it at CES. Instead, we used Bear's, (our Bear here) power amp with great success with the CTC Blowtorch at CES. Years later, we could use the CTC modified HCA-3500 with sonic success, and then we went on to the JC-1 which is even more developed.
John,
Wrong forum. Something like defence focused sites may be more interested in that side.
I am rather happy that I bid the whole military/industrial/political complex a "farewell and up yours with an RPG" decades ago, too much of this stuff is just too sick even for hardened cynic and rude boi like me...
That said I still like to occasionally reminisce about training warpac sniper rifles, blowing up stuff and other fun excercises. Great times, we baked our bread dough and fried our eggs on the Armour of the APC's in the summer (they had ABC protection but next to no aircon) and mixed medical alcohol with orange juice crystals and water for drinks.
Said in a "Tweety the Bird" Voice: "I lost more braincells that way!"
Ciao T
Wrong forum. Something like defence focused sites may be more interested in that side.
I am rather happy that I bid the whole military/industrial/political complex a "farewell and up yours with an RPG" decades ago, too much of this stuff is just too sick even for hardened cynic and rude boi like me...
That said I still like to occasionally reminisce about training warpac sniper rifles, blowing up stuff and other fun excercises. Great times, we baked our bread dough and fried our eggs on the Armour of the APC's in the summer (they had ABC protection but next to no aircon) and mixed medical alcohol with orange juice crystals and water for drinks.
Said in a "Tweety the Bird" Voice: "I lost more braincells that way!"
Ciao T
Most of us learned 'real quality design' by working on military projects on either side of the Cold War. I worked first on military designs, myself. Then NASA, then Audio, etc.
I almost got to work for GPTP, it was a company between Soviet science & engineering for military electronics. Thanks God I hesitated, otherwise it would be troublesome to get passport and go abroad, it is not good to know about future producton for military purposes...
Phrase from "Diamond Arm" movie, "As used to say my boss, now dead, 'I knew too much'" 😀
However, I was trained to operate tactial mobile anti-aircraft rocket system, (by the way, almost tube based), on light tank chassis. It was compusory in USSR, in institutes and universities, to get 2'nd education as an officer, for reserve in case of war. We learnt American, German, British, Japanese, Friench, etc... aircrafts, as targets of "Potential Enemy".
Phrase from "Diamond Arm" movie, "As used to say my boss, now dead, 'I knew too much'" 😀
However, I was trained to operate tactial mobile anti-aircraft rocket system, (by the way, almost tube based), on light tank chassis. It was compusory in USSR, in institutes and universities, to get 2'nd education as an officer, for reserve in case of war. We learnt American, German, British, Japanese, Friench, etc... aircrafts, as targets of "Potential Enemy".
Last edited:
Well, back to the amp schematics. What are the fundamental differences between the 3500 and the A-21? Sharp eyes will note fairly quickly.
Well, the input cascodes and the VAS's are MOSFETs instead of BJTs. I'm surprised that neither uses a cascoded VAS - there must be a reason...
Much thanks,
Chris
Actually a cascode VAS would be an advantage, and would lower the overall distortion.
One problem was the IRF 9610 originally used, because of its inherent nonlinearity. Later we switched to Harris 9610 mos fet.
There are a number of subtle, but important differences.
One problem was the IRF 9610 originally used, because of its inherent nonlinearity. Later we switched to Harris 9610 mos fet.
There are a number of subtle, but important differences.
John,
Other than the Hexfet Cascode and VAS (which I personally would have avoided seeing their parasitic capacitances and the way they behave with signal) I see a LOT of bypassing, that at first blush does not seem to have been fully evaluated (run the stuff on a network analyser or at least a decent Sim).
It looks a lot like what I see from certain types of chinese products these days, where the idea is that this sort of thing offers "audiophile" qualities, when in fact it often destroys them.
Of course, neither might be the root cause.
Cute dynamic biasing circuit BTW...
Ciao T
Other than the Hexfet Cascode and VAS (which I personally would have avoided seeing their parasitic capacitances and the way they behave with signal) I see a LOT of bypassing, that at first blush does not seem to have been fully evaluated (run the stuff on a network analyser or at least a decent Sim).
It looks a lot like what I see from certain types of chinese products these days, where the idea is that this sort of thing offers "audiophile" qualities, when in fact it often destroys them.
Of course, neither might be the root cause.
Cute dynamic biasing circuit BTW...
Ciao T
Bypassing is one of the FIRST things we change to improve the product. In fact, we removed quite a few parts from the 3500 to make it 'right'. I have never understood why the Taiwanese like to add so many bypass caps. The JC-1 has less bypass caps, but better quality ones.
Although i have a preference for symetrical differential topology
the two schematics discussed in theses last pages seems to me
somewhat at the edge..
Particularly , the use of jfets for the input differentials render
theses amps quite difficult to reproduce reliably unless one
manage to match them accurately enough , a care that
also must be taken for the two VAS devices...
the two schematics discussed in theses last pages seems to me
somewhat at the edge..
Particularly , the use of jfets for the input differentials render
theses amps quite difficult to reproduce reliably unless one
manage to match them accurately enough , a care that
also must be taken for the two VAS devices...
Hi,
Well, Parasound seems to have no problems to reproduce these Amplifiers reliably... I guess there are reasons why there are professionals and amateurs.
Ciao T
Well, Parasound seems to have no problems to reproduce these Amplifiers reliably... I guess there are reasons why there are professionals and amateurs.
Ciao T
The original input jfets are factory matched pairs, wahab. There is no advantage to bipolars in a comp. diff. input stage. Trust me, I developed both approaches more than 40 years ago. I did bipolars for the first 4 years WITH factory matched pairs, as well. In fact, jfets are easier to make work right because they don't have beta differences. Actually a reasonable difference in Idss of even a few ma between the N and P jfets is not very important to the circuit's operation.
- Status
- Not open for further replies.