Those are not opposites, so not exactly a bivalenced point of strict logic. Could be used to illustrate a point of contrast in natural language though. I wouldn't expect you not to give THD numbers if you customers want to see it. Wouldn't hurt for you give a disclaimer though. That might go a long ways to countering any appearance of disingenuity.
Don't know if you know of the former forum member, Sy? Venerated by some, but not necessarily by me. Don't know how you feel about him. Anyway, he sometimes liked quote Upton Sinclair as follows: 'It is difficult to get a man to understand something, when his salary depends on his not understanding it.'
https://www.diyaudio.com/community/...orch-preamplifier-part-ii.146693/post-4775574
I wonder if some of that might be applicable in this case.
Also, IMHO the the 'difficulty in getting a man to understanding something' in the above context doesn't necessarily have to be at a conscious level on his part. Someone might even make a strong argument that there is not any 'free will' involved. https://www.amazon.com/Determined-Science-Life-without-Free/dp/0525560971
Last edited:
Whatever, dude. You be you.
Should we get back to discussing test equipment for measuring common performance parameters of amplifiers as OP requested?
Tom
Should we get back to discussing test equipment for measuring common performance parameters of amplifiers as OP requested?
Tom
I think Tom is right, so is Mark.
Unless you know what you are measuring, and know how to improve it, the information is useless.
You measure an opamp, and find the result within spec then what will you do with the information?
Can you change the value?
You have to design the circuit, bearing the capabilities of the components in mind.
Unless you know what you are measuring, and know how to improve it, the information is useless.
You measure an opamp, and find the result within spec then what will you do with the information?
Can you change the value?
You have to design the circuit, bearing the capabilities of the components in mind.
The man does not have a proper DMM, so I think measuring and correcting amplifiers is something he should not attempt at this stage of learning.
And it is like a measuring instrument or a tool, you must know how to use it.
As an aside, a friend had a heart valve replaced in 2010, after a doctor checked his chest with a stethoscope, then put it under his ear, listened closely, and said leaky valve suspected, get it investigated.
So what did the stethoscope have to do with the diagnosis?
It was simply a good tool, was it not?
Tom runs a business making amplifiers, he worked in large companies before that, he is used to operating such equipment and interpreting the results, then doing something about it, if needed.
Which ability may be lacking in the OP.
Mark, please do not be offended, I know little about your abilities.
THD+N is a subject of much discussion, but most amps give decent results between 20 and 80% power, so measuring all the components, building an amp, measuring, improving, then whole cycle again is not realy a hobby, it is hard work and not worth the hassle for a small run of less than 10 pieces.
Leave it to the pros.
Tom has put a lot of work on his circuits, and his measured figures have to be verifiable, if a potential client needs to see them.
And those values should be independently verifiable at a competent measuring site anywhere in the world.
That is what makes claims genuine, not snake oil.
The other thing is that today with Class D, and digital sound, and lowered sales of analog sound systems, the effort in designing and building a totally new low noise design with good THD+N figures may be wasted, unless it is for personal satisfaction.
What is commercially available at the high end has been better than 100 dB since CDs came in the market, about 40 years back.
Now it may be 140 dB S/N ratio, but still, I feel not much can be achieved without going exotic, and the measuring equipment alone will be more expensive than the amp!
And it is like a measuring instrument or a tool, you must know how to use it.
As an aside, a friend had a heart valve replaced in 2010, after a doctor checked his chest with a stethoscope, then put it under his ear, listened closely, and said leaky valve suspected, get it investigated.
So what did the stethoscope have to do with the diagnosis?
It was simply a good tool, was it not?
Tom runs a business making amplifiers, he worked in large companies before that, he is used to operating such equipment and interpreting the results, then doing something about it, if needed.
Which ability may be lacking in the OP.
Mark, please do not be offended, I know little about your abilities.
THD+N is a subject of much discussion, but most amps give decent results between 20 and 80% power, so measuring all the components, building an amp, measuring, improving, then whole cycle again is not realy a hobby, it is hard work and not worth the hassle for a small run of less than 10 pieces.
Leave it to the pros.
Tom has put a lot of work on his circuits, and his measured figures have to be verifiable, if a potential client needs to see them.
And those values should be independently verifiable at a competent measuring site anywhere in the world.
That is what makes claims genuine, not snake oil.
The other thing is that today with Class D, and digital sound, and lowered sales of analog sound systems, the effort in designing and building a totally new low noise design with good THD+N figures may be wasted, unless it is for personal satisfaction.
What is commercially available at the high end has been better than 100 dB since CDs came in the market, about 40 years back.
Now it may be 140 dB S/N ratio, but still, I feel not much can be achieved without going exotic, and the measuring equipment alone will be more expensive than the amp!
Last edited:
True that. Although, it is possible that OP has a DMM but just didn't list it.
Yeah. If you really want to push the limits on THD measurements you'll either have to accept that you'll be limited to measurements at a few discrete frequencies/amplitudes or you'll have to spend some serious cash.
That said, you can get very, very far with a sound card, a precision oscillator, and a notch filter. Then add a DMM capable of measuring AC voltages precisely up to at least 100 kHz and you'll even be able to calibrate the setup. The HP 34401A, 34410A, and 3478A mentioned previously all fit that bill.
Tom
Agreed, Tom, thanks.
Then the question is what will the OP do with the information given by the equipment?
Is he / she capable of modifying or otherwise improving the design?
Otherwise, the expense and effort is a waste.
Then the question is what will the OP do with the information given by the equipment?
Is he / she capable of modifying or otherwise improving the design?
Otherwise, the expense and effort is a waste.
Actually, just a thought...maybe some old recording / production studio equipment which may suit this use may be going at a scrap price?
So this guy buys it, and says what do I do with it?
We have had some similar threads over the years...
So this guy buys it, and says what do I do with it?
We have had some similar threads over the years...
This was also part of the text: "..."the "total harmonic distortion" may be interpreted sufficiently, but it is always preferable to specify each harmonic separately."
I believe that it needs to be included as it puts the "total harmonic content is not a measure of the degree of distastefulness to the listener,......" into context.
The harmonic content is more important than the "total" harmonic distortion.
I believe that it needs to be included as it puts the "total harmonic content is not a measure of the degree of distastefulness to the listener,......" into context.
The harmonic content is more important than the "total" harmonic distortion.
I am not offended at all. Concerning my capabilities, I have a lifetime of hardware development experience that started with tube amps and now focuses on class-D. Being retired, it is my hobby to fiddle with private guitar amps and measuring stuff. My measurement setup is quite basic nonetheless complete, with a Rigol Scope, an DIY arbitrary waveform generator, lab power supplies, some USB-soundcards, DIY dummy loads, DIY attenuators. No AP, HP or Tektronix in my cellar man cave.
A demonstrably false statement. Given some distortion is perceptible, then clearly doubling the distortion will increase that perceived distortion, however you measure it. So without any doubt there is an unassailable (positive) correlation between measured THD and measured IMD and perceived distortion for any particular distortion profile.
Which ways to measure distortion are the most useful - that's an entirely different matter.
What you can infer is that driving THD or IMD down towards zero reduces any distortion to zero too, that's the maths of reality. The only function with zero THD and zero IMD is f(x) = ax
Hi guys. First thank you so much all of you.
Also, I am not wanting to define the listening quality by the measurements, I just want to see the numbers, because I am tired of spending days looking at the datasheets, then finding on the forum that they do not translate in real life (for example TI and Analog Devices opamp.... if you look at AD datasheet they are almost all better than TI opamps, but not in real life).
I will definitely buy the APU ADCiso, and victor oscillator. Especially if it compares with AP stuff.
Also I already have a multimeter but a cheap Handheld one (Nowhere near 6 digits), will it really help me having a better one ? A 4 1/2 digits bench one is it enough for measurements or not ? It is way cheaper than a 6 1/2 digits and 5 1/2. But I don't really care about cost, I just want to right tool to get the job done, not wanting to spend on something overkill (or not). So I can optimise the budget for other equipement.
I am also looking at buying "shield" are hammond diecast ALUMINIM boxes better, than a cheap empty STEEL Can of paint ?
Also, I am not wanting to define the listening quality by the measurements, I just want to see the numbers, because I am tired of spending days looking at the datasheets, then finding on the forum that they do not translate in real life (for example TI and Analog Devices opamp.... if you look at AD datasheet they are almost all better than TI opamps, but not in real life).
I will definitely buy the APU ADCiso, and victor oscillator. Especially if it compares with AP stuff.
Also I already have a multimeter but a cheap Handheld one (Nowhere near 6 digits), will it really help me having a better one ? A 4 1/2 digits bench one is it enough for measurements or not ? It is way cheaper than a 6 1/2 digits and 5 1/2. But I don't really care about cost, I just want to right tool to get the job done, not wanting to spend on something overkill (or not). So I can optimise the budget for other equipement.
I am also looking at buying "shield" are hammond diecast ALUMINIM boxes better, than a cheap empty STEEL Can of paint ?
Last edited:
Before buying high tech stuff it is helpful to do some numbers.
For instance the cheapest DVMs nowadays have an accuracy in the ballpark of 1%.
A audio level deviation of 1% corresponds a variation close to 0.1dB.
So, you will never ever need a 6 digit DVM for audio measurements.
I will encourage you to push the equipment already at your hands to its limits.
This means you have to learn use them effectively at first.
After reaching these limits, you may expand with more sophisticated stuff like the victor oscillator. T-filtering etc.
For instance the cheapest DVMs nowadays have an accuracy in the ballpark of 1%.
A audio level deviation of 1% corresponds a variation close to 0.1dB.
So, you will never ever need a 6 digit DVM for audio measurements.
I will encourage you to push the equipment already at your hands to its limits.
This means you have to learn use them effectively at first.
After reaching these limits, you may expand with more sophisticated stuff like the victor oscillator. T-filtering etc.
Even instruments you can make yourself like Bob Cordell’s distortion magnifier, which is a project I want to do to go along with my amber 3501 and I have a few of viktors oscillators too.
The benefits of a HP3478A or other more expensive models was its AC specs, 4-wire ohms for low resistance. if you have a scope that is usually good enough for most AC, esp if it has measurement capabilities.
As said run with what you have until you hit those limits then figure on what you need to improve upon.
The benefits of a HP3478A or other more expensive models was its AC specs, 4-wire ohms for low resistance. if you have a scope that is usually good enough for most AC, esp if it has measurement capabilities.
As said run with what you have until you hit those limits then figure on what you need to improve upon.
Last edited:
I think of it as a reasonable engineering approximation in the context of "degree of distastefulness to a listener." Besides, if you don't like Geddes, there is still Olive, who says:
Now, if we can move along given the OP has returned and explained his reasoning?
Last edited:
Okay. So is it that you feel somebody is not telling the truth in datasheets? I mean, since you are not equipped to measure yet, how are now determining, "not in real life"?
Do the diyaudio users who measured know why "not in real life"?
Maybe it would be more clear if you could direct us to an example or two so we can see what you are referring to?
Maybe it would be more clear if you could direct us to an example or two so we can see what you are referring to?
ADA4627-1 vs opa627 for one. Can we stick to the subject. Please.
My waveform generator is rated for :
Sine THD <0.8%(20Hz-20KHz@0dBm)
(Output : 20Vpp<10MHz@50ohm)
it is also 14bit.
Will it be better to use the soundcard 24bit, and "probably" lower THD ?
Thank you but I want to start directly with the measurements. Better fine tune my "professional process", than losing time with useless data. Better to get usable data from the start, and then refine the "good" process.
My waveform generator is rated for :
Sine THD <0.8%(20Hz-20KHz@0dBm)
(Output : 20Vpp<10MHz@50ohm)
it is also 14bit.
Will it be better to use the soundcard 24bit, and "probably" lower THD ?
Okay. Are you interested in being able to take this type of distortion measurement?
Like I said, you will measure something ready made, a chip.
Can you change its behaviour, in a way to improve its basic performance?
I do not think so, and most of the posters here will agree.
So you effort will be useless.
And you equipment will be a set of boy toys.
The voltages, gain and slew rates are well documented by the maker, and reference / modified designs are well known, the chips are at least 15 years old.
So whatever could be done is finished, and well known.
And:
https://www.diyaudio.com/community/...outperform-burr-browns-opa-627-opa627.150222/
So at least in this case, a search would have saved us all the trouble of advising you......if the information had been provided earlier.
Can you change its behaviour, in a way to improve its basic performance?
I do not think so, and most of the posters here will agree.
So you effort will be useless.
And you equipment will be a set of boy toys.
The voltages, gain and slew rates are well documented by the maker, and reference / modified designs are well known, the chips are at least 15 years old.
So whatever could be done is finished, and well known.
And:
https://www.diyaudio.com/community/...outperform-burr-browns-opa-627-opa627.150222/
So at least in this case, a search would have saved us all the trouble of advising you......if the information had been provided earlier.