Re: god i hate nad amps
Borrowed a couple of NAD monoblocks in the late 80's (not sure of the model) for listening tests. After a couple of days, boxed em back up. Definately much too noisey. Test over, man. Never did look at NAD again after that experience.
PCB checking. Don't you check the netlist after importing to the layout tool? Doublechecking works for me too.
sakis said:
Borrowed a couple of NAD monoblocks in the late 80's (not sure of the model) for listening tests. After a couple of days, boxed em back up. Definately much too noisey. Test over, man. Never did look at NAD again after that experience.
PCB checking. Don't you check the netlist after importing to the layout tool? Doublechecking works for me too.
the 2SD753's and their complements are discontinued. you might want to look at using MJ15024/5 outputs, as they are 250V devices. for your input stage, you would be better off with a +/-15V rail derived from a separate low voltage winding and use 3 terminal regulators. Fender uses a similar zener/resistor combination in some of their amps, and they use 10 watt resistors, and it's still a reliability nightmare. the zeners run hot and are prone to shorting, and burning up the 10 watt resistors when they do. that tends to leave a lot of scorched real estate on a board. with Fender's design, that also tends to latch the amp output to one of the rails, and has a tendency to take out output devices and voice coils simultaneously. like the schematic shown, protection circuits are nonexistent.
I agree with Bonsai, in the least redesign the input ltp, add another pair of output transistors and no need to use that high a voltage.
BTW the topology behind the amp is not bad, that vas will be pretty good, I am wondering who was first using that push pull type ltp, tom holman or whoever designed this amp. Ive built tom holman type amp, and find it to have monsterous good clean bass, overall a good amp.
Sakis if youre interested and not too much in a hurry, we could put heads together and redesign some parts of this amp and turn it into a interesting project.
BTW the topology behind the amp is not bad, that vas will be pretty good, I am wondering who was first using that push pull type ltp, tom holman or whoever designed this amp. Ive built tom holman type amp, and find it to have monsterous good clean bass, overall a good amp.
Sakis if youre interested and not too much in a hurry, we could put heads together and redesign some parts of this amp and turn it into a interesting project.
I got 170W into 8r0 from +-58Vdc using a 40Vac transformer.
Your 200W target will be easily met using a 45vac transformer.
There is absolutely no need to go to +-90Vdc for a 200W amplifier.
The economic part of that 40Vac decision is, it allows one to use 63Vdc capacitors, saving a lot of space and money.
Your 200W target will be easily met using a 45vac transformer.
There is absolutely no need to go to +-90Vdc for a 200W amplifier.
The economic part of that 40Vac decision is, it allows one to use 63Vdc capacitors, saving a lot of space and money.
guys thanks !!!!
thanks for all but i was actually making a relaxation exrecise yesterday so thats why i made this pcb .... its exact copy of the original schematic ( unless i made some mistake ) but designed for plastic transistors and has 4 more transistors per board ( so transitor wise it will be almost enough to operate wit 90+90 rails )
i mostly did it for fun ...but personaly i would never built an amp supposed to be for PA use ( this is what are we looking at here ) without VI limmiters on board .....
also i agree that from 90 volts to 15 is a long way to go ....so also there should be some improovement
but on the other hand if this was a working circuit comercially made and the only think i did is to add a few more transistors then with a very sophisticated protection circuit it might as well work like hell .....
far more simple circuits with by far more minimalistic topology work like hell and prooved to be perfect tools to make money and sound in small Pa applications ( i refer to the modified HILL chameleon clones iv been runing for more than a year now absolutelly problem free )
thanks though
Bonsai said:
homemodder said:
thanks for all but i was actually making a relaxation exrecise yesterday so thats why i made this pcb .... its exact copy of the original schematic ( unless i made some mistake ) but designed for plastic transistors and has 4 more transistors per board ( so transitor wise it will be almost enough to operate wit 90+90 rails )
i mostly did it for fun ...but personaly i would never built an amp supposed to be for PA use ( this is what are we looking at here ) without VI limmiters on board .....
also i agree that from 90 volts to 15 is a long way to go ....so also there should be some improovement
but on the other hand if this was a working circuit comercially made and the only think i did is to add a few more transistors then with a very sophisticated protection circuit it might as well work like hell .....
far more simple circuits with by far more minimalistic topology work like hell and prooved to be perfect tools to make money and sound in small Pa applications ( i refer to the modified HILL chameleon clones iv been runing for more than a year now absolutelly problem free )
thanks though
Andrew you are right, 90 volts is crazy, I know some audiophile friends that have first hand experience of of 70s 80s amps and they told me this was at the time a very good sounding piece of equipment, out of curiosity at what this amp may be able to achieve Im gonna run it in simulation but redesign the front ltp and use modern devices.
Sakis this is not a PA amp, but its pretty easy to incorporate VI limiters here, 2 extra trannies four resistors 2 diodes, extra pair of outputs. I will slap on a improved output stage, cfp ef which I prefer for PA and car amplifier use.
Nowadys there is not much diffrence between consumer grade hifi and pa amps, there are PA amps which are very well designed and will fit in perfectly in a home setting with comparable good sound.
Nowadys there is not much diffrence between consumer grade hifi and pa amps, there are PA amps which are very well designed and will fit in perfectly in a home setting with comparable good sound.
Hi,
were the 700 series amps Armstrong's first foray into SS output stages without an output transformer?
Or did that tribute belong to the 600 series amplifiers?
The Armstrong was a fairly good value, cheap British amplifier that just got into the HiFi category, when HiFi used to mean striving for High Fidelity.
I have no idea how reliable the Armstrongs were nor how long it took them to develop transformerless output stages. This could be a risky project.
were the 700 series amps Armstrong's first foray into SS output stages without an output transformer?
Or did that tribute belong to the 600 series amplifiers?
The Armstrong was a fairly good value, cheap British amplifier that just got into the HiFi category, when HiFi used to mean striving for High Fidelity.
I have no idea how reliable the Armstrongs were nor how long it took them to develop transformerless output stages. This could be a risky project.
The gentlemen I know are pensioner british citizens now living here, I see them sometimes as my girlfriend provides some medical care for them, they are in their late 70s, we didnt get into any details about amstrong amps, I only mentioned if they had ever heard of the 700 series amp by armstrong, to which the reply was immediately confirmed that it was a 200 watter that sounded darn good, The only other comment was that they sounded slighly warmer than the typical british sound, I dont know what that means though.
Andrew its a piece of cake just redesigning that front ltp, I dont like it at all, it spells disaster to me, but I like that vas, it works very well, I recently build prototype holman amp, slightly different but used cascoded fet input ltp, very nice amp, I m trying to find someone here that can measure accurate THD20. The sims shows 115db 20 khz performance at half power H2 dominant , if sim is out by 10 15 percent it is still very good. Ostripper is in love with his holman type amp as well, for some reason that topology gives awesome bass.
Andrew its a piece of cake just redesigning that front ltp, I dont like it at all, it spells disaster to me, but I like that vas, it works very well, I recently build prototype holman amp, slightly different but used cascoded fet input ltp, very nice amp, I m trying to find someone here that can measure accurate THD20. The sims shows 115db 20 khz performance at half power H2 dominant , if sim is out by 10 15 percent it is still very good. Ostripper is in love with his holman type amp as well, for some reason that topology gives awesome bass.
could be the heat dissipated by the 15v regulators????
actually the Apt-1 had a lot of innovative features in it. it had a darlington diff amp (LTP), a current mirrored cascode VAS, a baker clamp to prevent rail hang-up during clipping, and a triple darlington (triple EF) output stage, a very effective and reliable DC servo, and a protection circuit of which i have yet to see it's equal. they were truly the most "idiot proof" amps i've seen. and, yes, they did have awesome bass.
unclejed613, you right about the holman amp, its pretty simple but very well designed. The proto I built I changed the cascode to a randy sloan type, referenced to the output, Im not sure what the right english terminology is for this, more defined mids and treble without losing that stumping bass. Also I used a cascoded jfet input instead of the bjts, the holman is an old but very good design.
This armstrong uses the same topology but it has added ltp frontend and no cascode. Its openloop gain is higher but what I like about it is the wider bandwith by far, and its faster. Reminds me of the otala amps and what that designer tried to achieve, Im am owner of electrocompaniet monoblocks and they sound awesome. The current mirror I included in my simmed circuit, gives better balance of the push pull currents but not stricly needed, does nothing for distortion figures. Well if I strap a triple darlington on this the distortion figures would be scary, this is in simulation offcourse. I dont care much about clipping or protection circuits, I find them detrimental to the sound, If I want a 200 watt amp, Ill design it so it can handle 300w but Ill change the gain in accordance with preamp so it wont go far beyond 200. I see this is what armstrong have said about protection circuits too, I happen to agree.
This armstrong uses the same topology but it has added ltp frontend and no cascode. Its openloop gain is higher but what I like about it is the wider bandwith by far, and its faster. Reminds me of the otala amps and what that designer tried to achieve, Im am owner of electrocompaniet monoblocks and they sound awesome. The current mirror I included in my simmed circuit, gives better balance of the push pull currents but not stricly needed, does nothing for distortion figures. Well if I strap a triple darlington on this the distortion figures would be scary
, this is in simulation offcourse. I dont care much about clipping or protection circuits, I find them detrimental to the sound, If I want a 200 watt amp, Ill design it so it can handle 300w but Ill change the gain in accordance with preamp so it wont go far beyond 200. I see this is what armstrong have said about protection circuits too, I happen to agree.
actually the Baker clamp's only effect on the sound was to make light clipping undetectable sonically. and it probably saved a lot of users money, because the amp didn't do the "tweeter eater" thing when clipped. not that an amp should be run into clipping, but with music with a wide dynamic range (the intro to Boston's "Been So Long" comes to mind) it's almost inevitable.
the triple darlington used very fast predrivers and drivers, and the emitter loads did not return to the speaker rail. if you look at a schematic of the Apt output stage, you'll see what i mean. the beta on the output devices had to be below 45 or 50, to avoid common mode conduction, and this also improved the distortion figure a bit, since beta droop , while still there, was minimized as a matter of scale.
the triple darlington used very fast predrivers and drivers, and the emitter loads did not return to the speaker rail. if you look at a schematic of the Apt output stage, you'll see what i mean. the beta on the output devices had to be below 45 or 50, to avoid common mode conduction, and this also improved the distortion figure a bit, since beta droop , while still there, was minimized as a matter of scale.
The 600 series was also SS without output transformers
I've got a 600 series amp which sounds absolutely sublime (following a recap). If thats anything to go on then I would be very eager to attempt this build at some point. Only 200 or so units of the 700 series were ever sold, so it is a rarity.
Apparently the Armstrong 700 series was the companies break away from quality budget hifi to precision high quality (high price) gear, and due to this the company went bust. People weren't ready to pay that much for an Armstrong I guess when they could get a Quad or a Linn for the same price.
Having said that when I got in touch with one of the original designers (Mike Solomon) looking for schematics and pcb designs his initial response was "It's still working?!?!"
Funny thing this Armstrong amp. Is it built for something like 100R speakers or is one of the first attempts to make self oscillating amplifier? I think, it will oscillate so much that it will overheat at idle. Or will burn when goes to clip. What would happen if you hook up 4R speakers? You need to water cool it. Please don't build it, at least not without any caps to calm down oscillations......
Have fun
Have fun
wow, no compensation caps, in fact no bandwidth limiting of any kind in this schematic. looks like a good topology, just poorly executed... with the brute force zeners and the obvious lack of compensation, are we sure this isn't an expensive space heater? somebody mentioned water cooling..... liquid nitrogen is more like it....(waddya mean the transistors won't work below -40 degrees?)
Last edited:
High rail voltage with 2 pairs of output transistors to get 200-300W to drive real world speakers, why not?
I wonder why I haven't seen here an effort to use a cascoded output stage?? I don't understand amp design so I wonder why not?!
I hope that sakis will come up with successfull result so I can follow building this amp. But may be not, I don't have that many bipolar pairs, I expect only 2 pairs of output transistors
Why not? Because you can only dissipate some 50-60W per transistor if use mica insulators and have operating temperature of heat sink some 60 degrees celsius 8i don't know how much is this in farenheit). So there you get some 200W of heat dissipation and how effective is such amp? I suspect some 60%. So you can squeeze out just little over 200W. Things aren't so bad because w listen music not DC but such an amp will overheat sooner or later. Impendance of speaker isn't just simple 8E but can dramatically drop. Sometimes under 2E at poor constructed speakers. At +-90V juy will get some 60V output which gives you very high current peaks when Z drops. Sooner or later the outputs will fail. 100%.
Hmm liquid nitrogen is nice idea. Transistors will be superconductive and there will be no internal loses....it maybe will work. Lets try...he he
But if someone wants to see how looks output stage from that era then should find schematic of Phase Linear 700b. That amp have used +-100V. And runs hot as hell.
So, if someone think that is possible to build powerful and cheap amp(because there are only 4 transistors....), just because he find such scheme,than HE IS WRONG. Everything is possible on paper. Such amp will be very very unstable and unreliable. Especially if you install some modern transistors instead of venerable ,,heat dissipators,,
I know what happens when you just drop in new components in old circuits constructed with old materials. Try or believe.
Make a 1000cc car engine with 300hp.
Best regards
Taj
Hmm liquid nitrogen is nice idea. Transistors will be superconductive and there will be no internal loses....it maybe will work. Lets try...he he
But if someone wants to see how looks output stage from that era then should find schematic of Phase Linear 700b. That amp have used +-100V. And runs hot as hell.
So, if someone think that is possible to build powerful and cheap amp(because there are only 4 transistors....), just because he find such scheme,than HE IS WRONG. Everything is possible on paper. Such amp will be very very unstable and unreliable. Especially if you install some modern transistors instead of venerable ,,heat dissipators,,
I know what happens when you just drop in new components in old circuits constructed with old materials. Try or believe.
Make a 1000cc car engine with 300hp.
Best regards
Taj
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.