That should be true . There is a very strong argument for pre amps to be mini power amps . If the pre amp could almost drive speakers directly it is useful .
As I was saying before some power op amps look almost good enough to be universal devices . I think the market could use a proper device which is above NE5534 and would drive speakers . Dvv's boosters seem valid here .
5 is a minimum , 10 is what is taught . 3 can be OK . Less than 1 often happens between input pair and VAS . If it matters into the amp why not the VAS ? Transconductance ( 1 :1) , I don't believe in it although it seems to work with some amps . Using a high gain transistor for the VAS can do magic .
Best to have late trains than no trains . My grandfathers name was Beeching , the man who closed the railways was Dr Beeching ( cousin ) . No one liked him I'm told .
It's really all about current capability. If a preamp output can deliver some serious currents (e.g 200-300 mA, mine will do better than 2 Amps), and has a reasonably low output impedance, then parallel driving should not be a problem.
Tubes are cursed here. I made three headphone amps/preamps, one uses BJTs all over, the other has FET inputs and MOSFET outputs, and the third is an all tube design, featuring no global NFB. The tube job was designed by a good friend, who has been designing tube gear for the last 25 years or so, I wouldn't trust myself with tubes to make a unity gain buffer.
While it does music very well indeed, it is EXTREMELY load sensitive. Use 30 Ohms cans and the bass almost ain't gonna be there, but use 100 Ohm cans and it really comes on song.
God forbid that you should try driving cans, even 600 Ohms cans (Beyer Dynamic) in parallel with the power amp input (Marantz 170 DC, 82 Kohms, FET input) and the overall volume droops by about a half. PSU uses full regulation, 2 50 VA toroids for full true dual mono (same as the SS units).
On the other hand, both SS units will drive 8 Ohms in parallel with 27k of the HK 6550 with aplomb, level reduction (audibly) zero. Again, 2 50VA toroids, two regulated shunt PSUs.
I'll say it again and take the flack - FORGET op amp data sheets, they lie like dogs regarding available current output, and add a simple 2 stage current boost ciruit, using off the shelf devices. And it will rock!
Tubes are cursed here. I made three headphone amps/preamps, one uses BJTs all over, the other has FET inputs and MOSFET outputs, and the third is an all tube design, featuring no global NFB. The tube job was designed by a good friend, who has been designing tube gear for the last 25 years or so, I wouldn't trust myself with tubes to make a unity gain buffer.
While it does music very well indeed, it is EXTREMELY load sensitive. Use 30 Ohms cans and the bass almost ain't gonna be there, but use 100 Ohm cans and it really comes on song.
God forbid that you should try driving cans, even 600 Ohms cans (Beyer Dynamic) in parallel with the power amp input (Marantz 170 DC, 82 Kohms, FET input) and the overall volume droops by about a half. PSU uses full regulation, 2 50 VA toroids for full true dual mono (same as the SS units).
On the other hand, both SS units will drive 8 Ohms in parallel with 27k of the HK 6550 with aplomb, level reduction (audibly) zero. Again, 2 50VA toroids, two regulated shunt PSUs.
I'll say it again and take the flack - FORGET op amp data sheets, they lie like dogs regarding available current output, and add a simple 2 stage current boost ciruit, using off the shelf devices. And it will rock!
Last edited:
I think a good universal buffer with as small a number of components as possible would be worth building . It could be used to drive small and/or efficient speakers / cans even. It could be used with passive pre amps . The challenge would be sound with or without equal or better . It should always be at least as good or better . I reckon 80 % of the time better .
Douglas Self has built a NE5532 power amp . In principle it could be that with better chips . 10 V rms 3 R , job done .
Douglas Self has built a NE5532 power amp . In principle it could be that with better chips . 10 V rms 3 R , job done .
Well Nige, there are quite a few chip power amps around, I'm told, some delivering quite reasonable power outputs.
When you add an active current gain power stage to a passive preamp, as I see it, it's no longer a passive preamp, just different.
As for op amps as buffers, they can do the job with fyling colours. You have your pick of devices, thank the Lord, some will insist on FET input op amps, I'll insist on AD 829 because of its speed and low noise, etc, there's something for everybody.
What is a "simple buffer"? Highly arguable term. But once again, there's something for everybody. I have generally used three types of buffers. One uses 2 2SK170 FETs and is a variation of the Borbely theme. Another is fully complementary and uses 6 transistors, usually low noise BC 550/560 B types, but while it may appear to be complicated, in fact it's not, it's rather simple and compact. The third uses an AD 829 op amp in unity gain mode. All work well and basically have very little, if any, sound of their own.
I'm quite sure most of us here could tell a story of their own on buffers. The only type of buffer I have not tried is a single tube type, I know far too little about tubes to be able to design anything half decent. And in those days, I didn't know ol' Wave.
When you add an active current gain power stage to a passive preamp, as I see it, it's no longer a passive preamp, just different.
As for op amps as buffers, they can do the job with fyling colours. You have your pick of devices, thank the Lord, some will insist on FET input op amps, I'll insist on AD 829 because of its speed and low noise, etc, there's something for everybody.
What is a "simple buffer"? Highly arguable term. But once again, there's something for everybody. I have generally used three types of buffers. One uses 2 2SK170 FETs and is a variation of the Borbely theme. Another is fully complementary and uses 6 transistors, usually low noise BC 550/560 B types, but while it may appear to be complicated, in fact it's not, it's rather simple and compact. The third uses an AD 829 op amp in unity gain mode. All work well and basically have very little, if any, sound of their own.
I'm quite sure most of us here could tell a story of their own on buffers. The only type of buffer I have not tried is a single tube type, I know far too little about tubes to be able to design anything half decent. And in those days, I didn't know ol' Wave.
Dvv , my idea is a universal discreet op amp . One which for fun mostly adapt to the use as a small amplifier . It could be for tweeters . As the PCB is accessible some parts could be left out if just a simple buffer required . Range Rover tried to do this with a "one does all " car . Apart from no power take off ( so easy if they wanted to ) it is a near perfect concept , a car that thinks it is a tractor . Like a shopping car we have op amps if not requiring the last word in power output . I get into trouble for talking cars so analogy only this time .
Dvv you said I design by taking pieces out . I would with this . However it must be better than the ready made op amps ( more bias is the easiest route ) . Most of what we see in op amps is because it is always preferable to construct a transistor rather than capacitor or resistor . Given that why do it the simple way when a different world is available ? For example a complex current mirror is no longer a choice , it is the correct choice . Those transistors with >3 junctions , where do they exist outside of op amps ?
Dvv you said I design by taking pieces out . I would with this . However it must be better than the ready made op amps ( more bias is the easiest route ) . Most of what we see in op amps is because it is always preferable to construct a transistor rather than capacitor or resistor . Given that why do it the simple way when a different world is available ? For example a complex current mirror is no longer a choice , it is the correct choice . Those transistors with >3 junctions , where do they exist outside of op amps ?
More bias is meaningful only if you have an output stage capable of benefiting from more bias. And soon enough, you realize you might well need heat sinks - that's when simple starts being complex.
Before you know it, your output pair is dissipating more than 1W each, continuous, and they tend to get rather hot at even such relatively small power levels. That's if you want it to be cabale of driving low impedance lines as well, if not, then perhaps two or three small signal BJTs in series/parallel can still do the job.
My experience is that whenever somebody tries to make something adjustable for everybody, you get a jack of all trades and a master of none. It can be good, but it can never rival the best, because the best are highly specialized for one function. This not only means selecting the right trannie for the job, but also optimizing the whole circuit for just those trannies. No small feat.
In my view, you'd be best off designing a discrete op amp in fully complementary topology - you need 10 transistors per channel for this, as I see it. The payback is that you could be able to drive actual real world speakers if you rescale the driver and output BJT pairs. The only limitation is that the pinout of the trannies must be the same.
Even using graden variety transistors, like say BC 550/560 B for input stages, BC 546/556 Bfor the rest and BD 139/140, you would still be able to use PSU lines of +/- 30V.
Before you know it, your output pair is dissipating more than 1W each, continuous, and they tend to get rather hot at even such relatively small power levels. That's if you want it to be cabale of driving low impedance lines as well, if not, then perhaps two or three small signal BJTs in series/parallel can still do the job.
My experience is that whenever somebody tries to make something adjustable for everybody, you get a jack of all trades and a master of none. It can be good, but it can never rival the best, because the best are highly specialized for one function. This not only means selecting the right trannie for the job, but also optimizing the whole circuit for just those trannies. No small feat.
In my view, you'd be best off designing a discrete op amp in fully complementary topology - you need 10 transistors per channel for this, as I see it. The payback is that you could be able to drive actual real world speakers if you rescale the driver and output BJT pairs. The only limitation is that the pinout of the trannies must be the same.
Even using graden variety transistors, like say BC 550/560 B for input stages, BC 546/556 Bfor the rest and BD 139/140, you would still be able to use PSU lines of +/- 30V.
Last edited:
Well, if they are handy ...
Personally, I much prefer plastic packaging simply because it's so much more convenient. Also, for line level applications and as long as it's not a power amp, I think MJE 15030/15031 is all one could ever wish for. Very well behaved, good beta, all of 8 Amp current capability, 140V, 30 MHz, what else is there?
And because they are relatively small, there's is literally a vast selction of heat sinks, from modest and cheap to even very good, and still cheap.
If you want custom, PADA of Italy, one of Europe's largest heat sink manufacturers, will be glad to comply. Pick a profile, tell them how many holes you want where, they drill them, polish them and anodize at the end. You get a truly professional look. PADA dissipatori per Elettronica, DISSIPATORE di calore
Personally, I much prefer plastic packaging simply because it's so much more convenient. Also, for line level applications and as long as it's not a power amp, I think MJE 15030/15031 is all one could ever wish for. Very well behaved, good beta, all of 8 Amp current capability, 140V, 30 MHz, what else is there?
And because they are relatively small, there's is literally a vast selction of heat sinks, from modest and cheap to even very good, and still cheap.
If you want custom, PADA of Italy, one of Europe's largest heat sink manufacturers, will be glad to comply. Pick a profile, tell them how many holes you want where, they drill them, polish them and anodize at the end. You get a truly professional look. PADA dissipatori per Elettronica, DISSIPATORE di calore
Fankly Andrew, I don't see from the data sheets just how is '34 "better", "faster", etc than the '30, except by the fact that it has a higher voltage rating?
Both have voltage ratings far in excess of what is required for a line amp transistor.
Or, you can use the concotion service gents use. It only assumes the pot is not encapsulated, but then, if it is, nothing will work.
You need one empty little bottle for eye or nose drops, one with which you can drop only one drop somewhere. Open it up, wash it with water and let it dry.
Buy some vaseline in your local drug store and some medicinal alcohol. Put some vaseline, about your small finger half nail amount, into the bottle you washed. Put in about an equal amount of medicinal alcohol, Close and shake for about 10 minutes or so. The result will be a goo which just a bit more liquid than jelly.
Turn pot to one side and drop just one drop, let it find its way in. Then turn the pot to the maximum opposite position and drop in another single drop. Move the pot slowly from one extreme to the next for about a minute. Pause for a minute, then repeat the moving for another minute.
I guarantee that pot will not do anything it's not supposed to do for the next 3 years at least.
NOTE: Under no circumstances use gasolime and/or petrol instead of the medicinal alcohol, beuase they have a high content of oil which you definitely do NOT want.
You need one empty little bottle for eye or nose drops, one with which you can drop only one drop somewhere. Open it up, wash it with water and let it dry.
Buy some vaseline in your local drug store and some medicinal alcohol. Put some vaseline, about your small finger half nail amount, into the bottle you washed. Put in about an equal amount of medicinal alcohol, Close and shake for about 10 minutes or so. The result will be a goo which just a bit more liquid than jelly.
Turn pot to one side and drop just one drop, let it find its way in. Then turn the pot to the maximum opposite position and drop in another single drop. Move the pot slowly from one extreme to the next for about a minute. Pause for a minute, then repeat the moving for another minute.
I guarantee that pot will not do anything it's not supposed to do for the next 3 years at least.
NOTE: Under no circumstances use gasolime and/or petrol instead of the medicinal alcohol, beuase they have a high content of oil which you definitely do NOT want.
Well i have some 150proof white rum, medicinal to some .....
Merry xmas Gents ......
Could be the salvation of your pot, Wayne.
servisol,
SERVISOL SUPER 10 SWITCH AND CONTACT CLEANER - FREE P&P | eBay
As usual use at your own risk..however I have had great results..
If you want to kill your pot use WD40...its almost guaranteed to make it as unusable as possible...(Dirt Magnet)
I have even seen people drill a small hole 1mm in an encapsulated pot and use servisol...Why..because if its coming out then there is nothing to lose..guess what it worked..
Regards
M. Gregg
SERVISOL SUPER 10 SWITCH AND CONTACT CLEANER - FREE P&P | eBay
As usual use at your own risk..however I have had great results..
If you want to kill your pot use WD40...its almost guaranteed to make it as unusable as possible...(Dirt Magnet)
I have even seen people drill a small hole 1mm in an encapsulated pot and use servisol...Why..because if its coming out then there is nothing to lose..guess what it worked..
Regards
M. Gregg
Last edited:
WD 40 is very good for Quad 33 switches . Jeff Popple of Quad recommended it . He said they started using it by accident when no switch cleaner was available . I used it on my 33 many years ago and it has been trouble free ever since . I do suspect it is superior for that , Jeff thought so .
Happy New Year when it comes . Took my friend out to lunch on the 20 th , wanted to be certain it was before the end of the world . Fox Inn Ardley Oxon . Time Warp on steroids .
Happy New Year when it comes . Took my friend out to lunch on the 20 th , wanted to be certain it was before the end of the world . Fox Inn Ardley Oxon . Time Warp on steroids .
- Status
- Not open for further replies.