FWIW, in my recent experience, the Fairchild parts that are substitutes for Toshiba ones have seemed to be quite decent. I purchase them from Mouser as well.
Brad
What is wrong with magnetic core driver transistors?
I once used medium power RF transistors for mic input amps, could not find anything better.
Well, I still have many hundreds of them, however, they are extra low voltage, medium noise, lowish beta, and are not optimized for low current operation.
Of course, I designed with them in 1968, but soon I found other devices that had more linearity, higher voltage, and more power dissipation. I gave up on the 2N2222 type device by 1972. Gee that was only 40 years ago. '-)
Of course, I designed with them in 1968, but soon I found other devices that had more linearity, higher voltage, and more power dissipation. I gave up on the 2N2222 type device by 1972. Gee that was only 40 years ago. '-)
I received my order of on-semi from mouser and they look like fakes, have'nt chked them yet , but they do look it , hopefully not ....
Hi,
The diagram you show appears to come from a circuit simulator, not a real amplifier. Is this correct?
In which case you should know that usually simulators models of transistors do not include several material non-linearities that can produce results between 20db and 40dB better than real circuits?
Why don't you show how a REAL 100W lateral mosfet amp behaves, instead of simulations that are not very good models of reality?
Please stop presented simulations as if they where measurements of real devices, to do so is quite deceitful.
Ciao T
PS, here a REAL 100W Lateral Mosfet Amp based on the hitatchi schematics measured in stereophile with 20V P-P (that is at 10dB lower power than your simulation) into 8 Ohm:
The diagram you show appears to come from a circuit simulator, not a real amplifier. Is this correct?
In which case you should know that usually simulators models of transistors do not include several material non-linearities that can produce results between 20db and 40dB better than real circuits?
Why don't you show how a REAL 100W lateral mosfet amp behaves, instead of simulations that are not very good models of reality?
Please stop presented simulations as if they where measurements of real devices, to do so is quite deceitful.
Ciao T
PS, here a REAL 100W Lateral Mosfet Amp based on the hitatchi schematics measured in stereophile with 20V P-P (that is at 10dB lower power than your simulation) into 8 Ohm:
Hi,
No. It is decided by audibility of supposed distortions.
Please first show me speakers that offer -100dB or better THD at rated input power, then I worry about having to make Amplifiers with -100dB THD.
It is worth sometimes, as a reality-check to consider the noise-floor and distortion of recording microphones as well as the distortion and maximum SPL of Hifi and High End Speakers and to consider at what peak SPL's we listen and what peak SPL's are the microphones position.
It rather puts things into perspective, if I may say so...
Ciao T
No. It is decided by audibility of supposed distortions.
Please first show me speakers that offer -100dB or better THD at rated input power, then I worry about having to make Amplifiers with -100dB THD.
It is worth sometimes, as a reality-check to consider the noise-floor and distortion of recording microphones as well as the distortion and maximum SPL of Hifi and High End Speakers and to consider at what peak SPL's we listen and what peak SPL's are the microphones position.
It rather puts things into perspective, if I may say so...
Ciao T
These are the transistors E.B. used in the Hafler amps. They are nothing special that's for sure, other than high Vce and very cheap. I suspect the latter was the driving factor. This round I am repairing, not designing new. After I get back to where I started, then I will ues 30 years of hindsight and what I have learned here in the last few weeks to see if there are much better parts and then work out the changes they might require.
That's understandable. However for 'new' designs, the 2N2222 is pretty marginal. It really was essentially used as a 'driver' for old core memory used in computers 40+ years ago. IF you get a detailed data sheet, you can learn a lot.
This particular device is VERY well defined in the early Motorola data handbooks of the '60's and '70's. It is revealing in that it shows that it is designed for high current operation, mostly, with marginal beta and relatively low breakdown voltage. Operation below 1 ma should be frowned upon. 5-25 ma might be best.
As I have tried to convey, tvr, you should read up on these factors. Your background will make it relatively easy to 'come up to speed' if you have access to the right textbooks. Two come to mind: 'The Art of Electronics' by H&H, and Cordell's 'Designing Audio Power Amplifiers'. Have those two textbooks at hand, and 90% or more of your questions will be answered.
This particular device is VERY well defined in the early Motorola data handbooks of the '60's and '70's. It is revealing in that it shows that it is designed for high current operation, mostly, with marginal beta and relatively low breakdown voltage. Operation below 1 ma should be frowned upon. 5-25 ma might be best.
As I have tried to convey, tvr, you should read up on these factors. Your background will make it relatively easy to 'come up to speed' if you have access to the right textbooks. Two come to mind: 'The Art of Electronics' by H&H, and Cordell's 'Designing Audio Power Amplifiers'. Have those two textbooks at hand, and 90% or more of your questions will be answered.
What about the aging of semiconductors ?
I'm worried about decalibration of my old multibit DAC.
Do newly produced semiconductors have better silicium (wafer quality) than, say, the ones produced 20 years ago ? Measureable better (noise) ?
Btw., I (now corretly) measure about 0.2% THD at 96% clipping voltage of my power amp (at 1kHz)
and -100dB noise level above 100Hz (falling, -110dB above 1kHz).
Are these typical values for a good ss-amp ?
I'm worried about decalibration of my old multibit DAC.
Do newly produced semiconductors have better silicium (wafer quality) than, say, the ones produced 20 years ago ? Measureable better (noise) ?
Btw., I (now corretly) measure about 0.2% THD at 96% clipping voltage of my power amp (at 1kHz)
and -100dB noise level above 100Hz (falling, -110dB above 1kHz).
Are these typical values for a good ss-amp ?
Last edited:
Hi,
Well, the Philips TDA1541A from the late 1990's taiwan production run measure by far more consistent and indeed lower noisefloor than earlier examples. In our CD-Players late production TDA1541A measure better than early double crown TDA1541A (mostly I think because we eliminate several additional factors that relate certain tolerances in the DAC Chips.
Also, modern PNP Transistors on chip are much better, modern PNP's in general are getting quite good and complementary to NPN actually, compared to the early 80's...
I can do this pretty much with a tube Amp, using modest feedback (around 26dB enclosing an output stage with additional local feedback and feedback to the driver stage, input stage open loop) if I can dial back to maybe 85 or 90% of reasonably hard clipping voltage.
I have made quieter 300B SE Poweramp's, if that is re. rated output power, including "below 100Hz"... The tube amp referenced re. THD probably also managed this level of noise. I used to run it on 98dB/2.83V/1m sensitive speakers without noise being audible with your ear next to the cone...
I have measured expensive high end solid state amp's that measured worse than what you note as well...
Ciao T
Well, the Philips TDA1541A from the late 1990's taiwan production run measure by far more consistent and indeed lower noisefloor than earlier examples. In our CD-Players late production TDA1541A measure better than early double crown TDA1541A (mostly I think because we eliminate several additional factors that relate certain tolerances in the DAC Chips.
Also, modern PNP Transistors on chip are much better, modern PNP's in general are getting quite good and complementary to NPN actually, compared to the early 80's...
I can do this pretty much with a tube Amp, using modest feedback (around 26dB enclosing an output stage with additional local feedback and feedback to the driver stage, input stage open loop) if I can dial back to maybe 85 or 90% of reasonably hard clipping voltage.
I have made quieter 300B SE Poweramp's, if that is re. rated output power, including "below 100Hz"... The tube amp referenced re. THD probably also managed this level of noise. I used to run it on 98dB/2.83V/1m sensitive speakers without noise being audible with your ear next to the cone...
I have measured expensive high end solid state amp's that measured worse than what you note as well...
Ciao T
Hi,
T
PS, here a REAL 100W Lateral Mosfet Amp based on the hitatchi schematics measured in stereophile with 20V P-P (that is at 10dB lower power than your simulation) into 8 Ohm:
OK, can you explain what is going on at 20K on that graph ..... ?
Is that with your good ear .....?
John,
1 up.
Horrorwitz & Hill is a daunting work in scale and size and may kill someone if you drop it out of the window of an upper floor in a towerblock, but it has all the basic stuff at undergraduate and graduate levels with reams of references and detail. Do not even attempt to read it cover-2-cover, but have it handy to look up details.
Robert Cordell's book I like because even though he may not "believe" in certain concepts associated with audiophilia, he does not brush them off, gives them a fair hearing and treatment (sometimes coming up with surprising insights in the process, like on protection relays and fuses) and shows even for concepts he disagrees with personally (e.g. zero looped feedback systems) how they may be implemented.
Interestingly, I have also found www.wikipedia.com a good source for reminders of what I learned decades ago at Uni, had few occasion to use and hence not exactly forgot (no-one ever actually ever "forgets" any mental engram well into very early childhood [some may even argue past lives] - we merely loose the index link for our internal versions of Google's search), more like forgot where on my brains internal SD card it bleedin well is stored.
As much as I disagree with D.Self's stance on many things in audio, his books are an excellent source for a certain type of plodding, exhaustive research into small details (while sadly avoiding making any great leaps and thus turning the massive quantity of data into a superior level of quality) and that makes them valuable to me, they contain a lot of detailed data nicely graphed for about any common iteration you may conceive. Just take his conclusions "cum grano salis magnitudinem montis"...
I would also recommend Samuel Groner's commentary on Self's Amplifier design Handbook and Samuels in general, I have been able to learn a decent lot:
SG-Acoustics · Samuel Groner · Power Amplifiers
In addition, Nelson Pass has many excellent DIY Articles, which if read carefully expose much more than is directly obvious.
Jan Didden's site Jan Didden audio diy and other human frailties place plays host to wide range of material regarding amplifiers that are otherwise hard to come by. The links are bit hidden, the main one is "My Library", but there is a hidden Easter-egg link within the Library page that has the really cute stuff.
Past that there are several sites (which we cannot link here) that share schematics of past and present audio products with minimal if any regards for copyrights, however I find especially japanese gear of the last few years of the 70's and of the early 80's a treasure trove.
If you ever get interested in tube related stuff, there is a massive sticky thread with on-line resources in the tube section (that modestly obliges me to not I had the honour to kick off) under "on-line tube learning for newbies", or something to that effect.
I find that understanding tubes well makes it easier to "see" the problems of solid state design and to conceive workarounds.
Finally, Steve Eddy I believe hosts a copy of the previously "secret handshake and initiation into the brotherhood required for access" John Curl Collection Acrobat File (which I believe does deserve a by far distribution than it has received).
Sadly the precise post with the link escapes my pitiful abilities to use the diyaudio.com search engine and locate it, however I have a copy and will be happy to make it available, lest Steve is unable to again post the link or John objects - send me a PM.
Ciao T
1 up.
Horrorwitz & Hill is a daunting work in scale and size and may kill someone if you drop it out of the window of an upper floor in a towerblock, but it has all the basic stuff at undergraduate and graduate levels with reams of references and detail. Do not even attempt to read it cover-2-cover, but have it handy to look up details.
Robert Cordell's book I like because even though he may not "believe" in certain concepts associated with audiophilia, he does not brush them off, gives them a fair hearing and treatment (sometimes coming up with surprising insights in the process, like on protection relays and fuses) and shows even for concepts he disagrees with personally (e.g. zero looped feedback systems) how they may be implemented.
Interestingly, I have also found www.wikipedia.com a good source for reminders of what I learned decades ago at Uni, had few occasion to use and hence not exactly forgot (no-one ever actually ever "forgets" any mental engram well into very early childhood [some may even argue past lives] - we merely loose the index link for our internal versions of Google's search), more like forgot where on my brains internal SD card it bleedin well is stored.
As much as I disagree with D.Self's stance on many things in audio, his books are an excellent source for a certain type of plodding, exhaustive research into small details (while sadly avoiding making any great leaps and thus turning the massive quantity of data into a superior level of quality) and that makes them valuable to me, they contain a lot of detailed data nicely graphed for about any common iteration you may conceive. Just take his conclusions "cum grano salis magnitudinem montis"...
I would also recommend Samuel Groner's commentary on Self's Amplifier design Handbook and Samuels in general, I have been able to learn a decent lot:
SG-Acoustics · Samuel Groner · Power Amplifiers
In addition, Nelson Pass has many excellent DIY Articles, which if read carefully expose much more than is directly obvious.
Jan Didden's site Jan Didden audio diy and other human frailties place plays host to wide range of material regarding amplifiers that are otherwise hard to come by. The links are bit hidden, the main one is "My Library", but there is a hidden Easter-egg link within the Library page that has the really cute stuff.
Past that there are several sites (which we cannot link here) that share schematics of past and present audio products with minimal if any regards for copyrights, however I find especially japanese gear of the last few years of the 70's and of the early 80's a treasure trove.
If you ever get interested in tube related stuff, there is a massive sticky thread with on-line resources in the tube section (that modestly obliges me to not I had the honour to kick off) under "on-line tube learning for newbies", or something to that effect.
I find that understanding tubes well makes it easier to "see" the problems of solid state design and to conceive workarounds.
Finally, Steve Eddy I believe hosts a copy of the previously "secret handshake and initiation into the brotherhood required for access" John Curl Collection Acrobat File (which I believe does deserve a by far distribution than it has received).
Sadly the precise post with the link escapes my pitiful abilities to use the diyaudio.com search engine and locate it, however I have a copy and will be happy to make it available, lest Steve is unable to again post the link or John objects - send me a PM.
Ciao T
Hi,
Yes. I can.
I have two good ears, actually.
My biggest problem is the occasional of positive feedback in the ears acoustic amplification system (some call it tinnitus) and a deep notch around 15KHz which I ascribe to my live & loud sound days (compression drivers overloading).
As bad as they may be, these ears (genuine MKI issue never upgraded though melticously clean) still serve adequately for the detection of hum and hiss and other audible phenomenae...
My best (single ended directly heated) tube amp designs have around 30uV of UNWEIGHTED noise, usually mains or low mains harmonics, less if we exclude these or use A-Weighting.
Ciao T
Yes. I can.
Is that with your good ear .....?
I have two good ears, actually.
My biggest problem is the occasional of positive feedback in the ears acoustic amplification system (some call it tinnitus) and a deep notch around 15KHz which I ascribe to my live & loud sound days (compression drivers overloading).
As bad as they may be, these ears (genuine MKI issue never upgraded though melticously clean) still serve adequately for the detection of hum and hiss and other audible phenomenae...
My best (single ended directly heated) tube amp designs have around 30uV of UNWEIGHTED noise, usually mains or low mains harmonics, less if we exclude these or use A-Weighting.
Ciao T
- Status
- Not open for further replies.