You know, Scott, you also imply that these tests will find something significant with the CTC Blowtorch, yet there is little or no reason to presume anything useful would be found. Therefore, I am COMMANDED to run these tests by people who apparently own no test equipment of their own, for their own edification, or perhaps just to 'hassle' me. Is this really going to serve a useful purpose?
John I asked on your behalf that folks just back off a liitle. This is not a subtle issue, Nelson's and other low feedback amps measured very bad on Ron's tasts, maybe we should back off and think about all this.
I didn't say (as I recall) that I would make this test. I SAID that I COULD DO this test. However, on second thought, it is a lot of hassle to set up and run the Hirata box, and Ron Quan is 50 miles away, and I can't drive that distance due to my vision problems.
ALSO, and even more important, especially to everyone here:
It is IMPORTANT to know about these 'exotic tests' like the Hirata and PIM, in order to AVOID designing a circuit that MIGHT generate SIGNIFICANT Hirata or PIM distortion, BECAUSE I have some idea as to what to avoid to not generate these 'exotic distortions'.
Is it NOW clear enough?
ALSO, and even more important, especially to everyone here:
It is IMPORTANT to know about these 'exotic tests' like the Hirata and PIM, in order to AVOID designing a circuit that MIGHT generate SIGNIFICANT Hirata or PIM distortion, BECAUSE I have some idea as to what to avoid to not generate these 'exotic distortions'.
Is it NOW clear enough?
Recently, I have be making some STANDARD tests with one of my Parasound power amps. I had initially hoped that this amp would replace the slightly aged Parasound prototype that I have been using for the last 10 years, off and on.
The input stage and the effective power into 8 ohms is about the same for either amp, the old prototype or the new model. However, the output stage of the prototype has 8 output pairs per channel and the other newer amp has only 4 pairs of output devices. Will this make a significant difference? Well, the new amp was disappointing into input loads of 4 ohms or less, AND my WATT's are 4 ohms or less. More devices, coupled with higher quiescent current is GOOD!
I used an HP339 distortion analyzer with an HP 3563 FFT analyzer with 100X sampling. With this test, I can easily go to -120dB and even -130dB when measuring the higher harmonics 5th-11th. I found too much higher order distortion at the class A-AB transition for me to be comfortable with.
The input stage and the effective power into 8 ohms is about the same for either amp, the old prototype or the new model. However, the output stage of the prototype has 8 output pairs per channel and the other newer amp has only 4 pairs of output devices. Will this make a significant difference? Well, the new amp was disappointing into input loads of 4 ohms or less, AND my WATT's are 4 ohms or less. More devices, coupled with higher quiescent current is GOOD!
I used an HP339 distortion analyzer with an HP 3563 FFT analyzer with 100X sampling. With this test, I can easily go to -120dB and even -130dB when measuring the higher harmonics 5th-11th. I found too much higher order distortion at the class A-AB transition for me to be comfortable with.
Can you precisely understand my ENGLISH LANGUAGE statements, Esperato?
First, "we should understand what the Hirata test is:" Do you even know what the Hirata test is precisely?
Second, "I have been asked to do a Hirata test" This does not mean that I have CONSENTED to do a Hirata test, and it would be useless to do so with MY designs, because I know what to prevent, to avoid Hirata distortion.
First, "we should understand what the Hirata test is:" Do you even know what the Hirata test is precisely?
Second, "I have been asked to do a Hirata test" This does not mean that I have CONSENTED to do a Hirata test, and it would be useless to do so with MY designs, because I know what to prevent, to avoid Hirata distortion.
I don't know what happened to my reply, but 'test' in English means two similar things.
1. It means a quiz or written exam given by a teacher.
2. It means applying test equipment to measure something.
3. The Hirata 'test' is a test procedure
4. 'Test on Friday' means a written exam, as in a college class.
5. They are not the same meaning, and this is what has lead to this misunderstanding.
1. It means a quiz or written exam given by a teacher.
2. It means applying test equipment to measure something.
3. The Hirata 'test' is a test procedure
4. 'Test on Friday' means a written exam, as in a college class.
5. They are not the same meaning, and this is what has lead to this misunderstanding.
English is not my language. But i'm a specialist in bad faith. Nice try.
Note, Carl, that i did not made any comments, just "your words".
Last edited:
but there is no such thing as "Hirata Distoriton" as a fundamental signal/circuit characteristic - the papers clearly relate the numbers from the proposed test waveforms to simple static nonlinearities
Cabot shows the same nonlinearities can be detected with AP's FastTest
and as usual the Hirata test waveforms can be well approximated by Fourier series
Cabot shows the same nonlinearities can be detected with AP's FastTest
and as usual the Hirata test waveforms can be well approximated by Fourier series
Esperado, I take exception to your attitude about this. I just set you straight about what I said, and I had no idea it would be 'ambiguous', such that you thought that I was going to run the Hirata test on Friday. No, I was going to test YOU on the Hirata test, Friday, instead. It was meant to be a modest joke, as I was playing the role as a college professor, and was only repeating what I remembered being told, myself.
it would be foolish not to - use all the tools, each where it works best
nice try John, but rather than making a “hit”, you just sound silly
it certainly is a fast way to see how much of the numbers from the real hardware are explained by the models
kinda the basis of EE - comparing models in our heads, on paper, or a computer screen - can't find the "X" factor if you have no clue what is already known, how the data relates to some model of the system
as EE/circuit designers our technology relies on simplified models - some so simple we can wrap our heads around them, write equations down on paper
we can use computers to work more complicated equations, larger systems of interacting equations, we can tune the models, equations to agree better with our measurements - or discover holes in the models
but our understanding, advancing tech relies on these models, heck we even have to have models of the whole measurement process to interpret the range of validity of the measurements themselves
even fuzzy heuristic models of the word that arise from unstructured observation are still models, which to be useful have to make predictions – that can be tested against real world
nice try John, but rather than making a “hit”, you just sound silly
it certainly is a fast way to see how much of the numbers from the real hardware are explained by the models
kinda the basis of EE - comparing models in our heads, on paper, or a computer screen - can't find the "X" factor if you have no clue what is already known, how the data relates to some model of the system
as EE/circuit designers our technology relies on simplified models - some so simple we can wrap our heads around them, write equations down on paper
we can use computers to work more complicated equations, larger systems of interacting equations, we can tune the models, equations to agree better with our measurements - or discover holes in the models
but our understanding, advancing tech relies on these models, heck we even have to have models of the whole measurement process to interpret the range of validity of the measurements themselves
even fuzzy heuristic models of the word that arise from unstructured observation are still models, which to be useful have to make predictions – that can be tested against real world
Last edited:
Mr. Wurcer, excuse me but I can't let JC get away with this.
- JC, you have a Hirata box.
- You've been at great pains to educate us on the Hirata test and how important it is.
- You tell us you will use your Hirata box to test Blowtorch on a Fri.
- The next week, instead of posting your results, you pontificate at length about how us unwashed masses won't be able to understand the results.
You are now claiming from your 50 yrs as a (pseudo?) guru, Blowtorch is incapable of generating Hirata or Quan's PIM so NO TESTING NEEDS TO BE CARRIED OUT ... enumerating loadsa reasons for not doing so.
My guess is you DID do Hirata on Blowtorch and found it wanting.
I'm now making a claim from my 50 yrs as a pseudo guru. From examining the circuit and from information on this thread, Blowtorch will perform worse on either of Ron Quan's tests than what he measured for 4558.
I may be completely wrong. A friend demonstrated a flying pig to me last Thu.
But you can settle the issue. IF Blowtorch is as good as you say, it will blitz both Hirata & Quan. Do the tests & put the doubters to shame.
If it doesn't, all it shows is incompetence. Nothing wrong with that. I'm sure even Scott admits to lapses of incompetence. I certainly do as a pseudo guru.
But to pretend a test wasn't carried out cos the results show Blowtorch in a bad light is far more serious than incompetence. Words like Christophe's 'fakir' come to mind.
Its a good thing Blowtorch is no longer sold cos then words like 'fraud' & 'misrepresentation', with legal repercussions rear their ugly head.
[/rant]
- Status
- Not open for further replies.