Lars Clausen said:
Agreed, op-amps wouldn't be my first choice but with a limited budget a cost saving here would make more money available for the PSU that might bring about larger overall gains.
Perhaps a guideline budget for a complete (non chip) stereo amplifier including case and PSU should be decided on. The best sonic performance for minimum cost.
Any advance on 75 euro?
Low Cost Enclosures
For me the cheapest ways of getting a good enclosure have been:
A- Suplus comm equipment. There is a surpus store near me that has racks of obsolete switches and routers. The enclosure are typically very soundly made. I have bought them fro US$1 to US$20. The trick is to watch for them and buy when they are available even if you don't need one at the moment. Generally I just remove the insides and throw them out but sometimes fuse blocks, EIC modules and so on can be used. Since they usualy have a switch mode power supply there isn't much there but I hang onto the the transformers just in case.
B- Old audio equipment. You need to be selective. Some enclosures are mostly plastic which isn't too suitable. Go for the all metal ones. I recently found a PA amplifier (US$5) that included a shielded E-I transformer. Industrial strength constructuion. The voltage on the secondaries indicates I 65-70W amp is possible. The VA rating is a guess (350VA) based on the mains fuses. There is enough space available in the bax that I may get bold and try to build an integrated amp.
The cheapest heatsinks I've used were constructed by bolting many surplus CPU heatsinks (US$0.25/each) to one side of a piece of aluminum bar stock. These were the kind with a forest of long rods sticking up from them. Lots of surface area. Very sci-fi/mad scientist looking! The important step seemed to be using some fine sandpaper on the barstock to make it very flat then bolting them very tight. If one wanted it to look nice, tapping the holes rather than using nuts would work. Once again, I grabed them when I saw them not waiting for when I had a definite need.
For me the cheapest ways of getting a good enclosure have been:
A- Suplus comm equipment. There is a surpus store near me that has racks of obsolete switches and routers. The enclosure are typically very soundly made. I have bought them fro US$1 to US$20. The trick is to watch for them and buy when they are available even if you don't need one at the moment. Generally I just remove the insides and throw them out but sometimes fuse blocks, EIC modules and so on can be used. Since they usualy have a switch mode power supply there isn't much there but I hang onto the the transformers just in case.
B- Old audio equipment. You need to be selective. Some enclosures are mostly plastic which isn't too suitable. Go for the all metal ones. I recently found a PA amplifier (US$5) that included a shielded E-I transformer. Industrial strength constructuion. The voltage on the secondaries indicates I 65-70W amp is possible. The VA rating is a guess (350VA) based on the mains fuses. There is enough space available in the bax that I may get bold and try to build an integrated amp.
The cheapest heatsinks I've used were constructed by bolting many surplus CPU heatsinks (US$0.25/each) to one side of a piece of aluminum bar stock. These were the kind with a forest of long rods sticking up from them. Lots of surface area. Very sci-fi/mad scientist looking! The important step seemed to be using some fine sandpaper on the barstock to make it very flat then bolting them very tight. If one wanted it to look nice, tapping the holes rather than using nuts would work. Once again, I grabed them when I saw them not waiting for when I had a definite need.
I've used the "MIR method" before, although I did it for looks since I mounted all my parts on a piece (3/8" I think) of wood I had found lying around in a closet, I think it was teak, looked like it. Since I didn't have enough wood for anything else I used it that way. Since then I've used wood that way a couple of times, the resuting looks makes the inards of an amp or preamp (I used it for preamps and a phono-preamp) very "homey" and "cozzy".
Maybe it's just me, but I like it, especially since I'm always fidnling and often leave the chasis open for the inards to be seen.
Maybe it's just me, but I like it, especially since I'm always fidnling and often leave the chasis open for the inards to be seen.
Hi Lars,
A very interesting design, thank you for posting it. It should sound very good!
However, what is even more impressive is your willingness to discuss your design in this forum without rancour or ego.
A few questions:
1. VERY low input stage gain, and furthermore low gain with massive local feedback at the VAS stage. Are you absolutely convinced that low feedback SS PP amplifiers sound better? What is the damping factor, and what is the impression of power from this amp? High feedback amplifiers, properly done, can sound far more powerful than their watts indicate. I'd estimate no more than about 40dB of feedback in this design.
2. Can you be absolutely sure that T8, the emitter follower drive to the VAS, is necessary? I tried it and eventually dismissed it as not worth the trouble. However, YMMV, and obviously does, so there must be a reason.......
3. There appears no symmetry of diff pair input impedances; this has repercussions for offset control. Any problems here?
4. What is the pole of this amp, Lars? Your feedback scheme is certainly interesting, possibly set up for highly capacitive loading. Is the slew rate symmetrical?
5. R23 is quite low at 2R2; normally this is around 10R. Is this to ensure stability into highly capacitive loads, or merely to improve short term transient stability?
I really like your use of a bipolar driver stage with the output mosfets. This is extremely intelligent design; that gate is not well handled direct by most voltage amplifiers. And the charge suckout network is a nice touch.
Once again, thank you. Impressive circuit. There are some common points with the AKSA, but although topologies are similar, the philosophy is different; the AKSA is a high feedback design with a very different feedback regime (also split, however).
Cheers,
Hugh
A very interesting design, thank you for posting it. It should sound very good!
However, what is even more impressive is your willingness to discuss your design in this forum without rancour or ego.
A few questions:
1. VERY low input stage gain, and furthermore low gain with massive local feedback at the VAS stage. Are you absolutely convinced that low feedback SS PP amplifiers sound better? What is the damping factor, and what is the impression of power from this amp? High feedback amplifiers, properly done, can sound far more powerful than their watts indicate. I'd estimate no more than about 40dB of feedback in this design.
2. Can you be absolutely sure that T8, the emitter follower drive to the VAS, is necessary? I tried it and eventually dismissed it as not worth the trouble. However, YMMV, and obviously does, so there must be a reason.......
3. There appears no symmetry of diff pair input impedances; this has repercussions for offset control. Any problems here?
4. What is the pole of this amp, Lars? Your feedback scheme is certainly interesting, possibly set up for highly capacitive loading. Is the slew rate symmetrical?
5. R23 is quite low at 2R2; normally this is around 10R. Is this to ensure stability into highly capacitive loads, or merely to improve short term transient stability?
I really like your use of a bipolar driver stage with the output mosfets. This is extremely intelligent design; that gate is not well handled direct by most voltage amplifiers. And the charge suckout network is a nice touch.
Once again, thank you. Impressive circuit. There are some common points with the AKSA, but although topologies are similar, the philosophy is different; the AKSA is a high feedback design with a very different feedback regime (also split, however).
Cheers,
Hugh
Club of " poor boys " from " rich coutrys "
Guys, your ideal is to made " superamp " with price like one hamburger and in my opinion it is now "sport for you ". Certainly exist here exceptions ( Per - Anders, Jan Dupont and several another guys ), but most take this way. I am asking why ? Are you realy poor ? Is here any rich boy, which have no problem with transistor costed 2,50 buck instead 2,00 ?
Guys, your ideal is to made " superamp " with price like one hamburger and in my opinion it is now "sport for you ". Certainly exist here exceptions ( Per - Anders, Jan Dupont and several another guys ), but most take this way. I am asking why ? Are you realy poor ? Is here any rich boy, which have no problem with transistor costed 2,50 buck instead 2,00 ?
Lars Clausen said:
thanks, Lars. Nice to see another IRF540/9540 based design (I have a drawerfull of them).
I have done a similar design (but is more based on Nelson's MOSFET citation 12, with a T-driver stage like yours) utilizing the IRF devices. It simulated well but I am still building it so it is too early to see how it will actually sound. It is even simpler than yours.
I think it is a smart idea to use CCS on the tails of the differential pair as you can change rail voltages without being concerned about current going through the pairs.
T8 is quite unique and I have never seen it before. Why is it designed such a way?
Is this thread soly limited to discreet designs, or can a chip amp design be submited here?
I just ask, as I have (as with many others) been building a clone, and I heve to say that it is not costing me very much to do at all. FIstly, the main component for the system (the LM3875) I got free from national (which is ok, if you are just making one for yourself), the rest of the electronic parts, are just standard quality (1%metal film resistors, Jamicon caps, and a few large non-polars for decopling) which must have cost me no more than £4.00 (including all the parts for the power supply - not the transformer).
I have also made a pcb for this, which is certinaly smaller than the one for the alph-x group order, so if it were mass produced should cost less then the $5 they are asking.
The other parts I have used for this have also been fairly chhep, with the transformer comming in at a whole £5 (160VA, 18 - 0 - 18 from ebay), a cast aluminium enclosuer, £2.00 (ebay) and two large pentium II pasive heatsinks (about 1C/W, for £0.50 each on ebay).
So, in total I think the whole am will come in at about £40 before I paint the case, which will probably cost me another £20 for the proper paints and primers (seems stupid that the most expensive part of the amp will be the paint
), and it should still sound pretty good.
So, I'm not too sure how you could get much cheeper than this by using discreet components, unless you are willing to build the amp up on veroboard or something similar, that you gotten cheep and with components tat you already heve (is that cheep, or is that just forgetting that you have already payed for the bits?)
I just ask, as I have (as with many others) been building a clone, and I heve to say that it is not costing me very much to do at all. FIstly, the main component for the system (the LM3875) I got free from national (which is ok, if you are just making one for yourself), the rest of the electronic parts, are just standard quality (1%metal film resistors, Jamicon caps, and a few large non-polars for decopling) which must have cost me no more than £4.00 (including all the parts for the power supply - not the transformer).
I have also made a pcb for this, which is certinaly smaller than the one for the alph-x group order, so if it were mass produced should cost less then the $5 they are asking.
The other parts I have used for this have also been fairly chhep, with the transformer comming in at a whole £5 (160VA, 18 - 0 - 18 from ebay), a cast aluminium enclosuer, £2.00 (ebay) and two large pentium II pasive heatsinks (about 1C/W, for £0.50 each on ebay).
So, in total I think the whole am will come in at about £40 before I paint the case, which will probably cost me another £20 for the proper paints and primers (seems stupid that the most expensive part of the amp will be the paint
), and it should still sound pretty good. So, I'm not too sure how you could get much cheeper than this by using discreet components, unless you are willing to build the amp up on veroboard or something similar, that you gotten cheep and with components tat you already heve (is that cheep, or is that just forgetting that you have already payed for the bits?)
Well, that's fine but I wasn't too sure because of the origins of this thread it self, but if people want to keep it all discreet that's fine by me (I might even learn something). As, it mihgt be an interesting project to try out if it ever get to a final kit/design as I have always assumed that when I do finaly take the time to do a proper discreet design, I was going to go all out and build something as well as I can, seen as how much extra time it would take in designing and bulding it in the first place.
One thing that I would like to say, which i don't think has been mentioned here yet, which may be an interesting option (I haven't seen it in too many designs or kits out on the net) would be to create a design using SMD parts, as these would allow for a much smaller PCB, and if these are going to be produced, a group buy of SMD parts (resistors, caps and transistors) would probably only add a few cents to the cost if enough people joined in.
Hi,
Just when I was looking forward to seeing a Hugh Dean balanced valve preamp...Deep sigh...
Guess he's forgotten all about it....................
O.K, O.K...discretion guaranteed....
Mind you, with a little more persuasion I'd end up with solid state delinquents such as Peter Daniel and Jam in the Tubes forum...
We could do alot worse...
Kidding folks,
Just when I was looking forward to seeing a Hugh Dean balanced valve preamp...Deep sigh...
Guess he's forgotten all about it....................
O.K, O.K...discretion guaranteed....
Mind you, with a little more persuasion I'd end up with solid state delinquents such as Peter Daniel and Jam in the Tubes forum...
We could do alot worse...
Kidding folks,
Paul,
Hilarious!!!
It was critiqued by the Moonee Ponds Photographic Club, an august institution founded in the 1880s who scoff at digital technology, and consigned to the X-files.
Tragic really. My 15 minutes of fame only lasted about ten seconds. And I thought my ensemble, not to mention my myriad of power amplifier creations, was rather fetching........
Cheers,
Hugh
Hilarious!!!
It was critiqued by the Moonee Ponds Photographic Club, an august institution founded in the 1880s who scoff at digital technology, and consigned to the X-files.
Tragic really. My 15 minutes of fame only lasted about ten seconds. And I thought my ensemble, not to mention my myriad of power amplifier creations, was rather fetching........
Cheers,
Hugh
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.