| lumanauw |
The latest Zen is Zen ver 7, common inductance loading. Mr. Pass said that this amp cannot be trimmed by matching voltage or matching current, it has to be done with distortion analyzer.
He said that we can also built (if cannot buy one), like the IM analyzer (Audio Express).
Does anyone has copy of that article? Or have other DIY distortion meter that can be used for trimming this amp? |
|
|
| lumanauw |
| Mr. Pass said something about notch filter. What is notch filter, and how can it be used to tune ZV7? |
|
|
| mrothacher |
A notch filter is tuned to filter out the fundamental frequency of interest so that what you're left with is harmonic distortion and noise. You could observe the filtered signal on an oscilloscope or use an audio voltmeter.
You can construct a notch filter of your own.
Try:
www.sound.westhost.com |
|
|
| Nelson Pass |
OK Fred, here's your opportunity to play educator. I know you
can do it for us. :cool:
Lacking that, build one - there's enough information out there,
and it doesn't have to be particularly good. 1% scale will serve. |
|
|
| mrothacher |
NP, Fred: I hope my desciption was accurate. I'm working on a "Poor Guy's" Audio Precision test rig. It's a hardware / software design. Would you guys be willing to critique the design of a fledgling like me? Maybe even compare it to the real thing? (NP - I know you keep one at you bedside!):cool:
Thanks
Mike |
|
|
| banana |
Rightmark Audio Analyzer
Have been playing with it recently. It's free and it's interesting.
It has been suggested in this forum before. But I still would like to hear more commend about it. Any more RMAA users here? |
|
|
| Nelson Pass |
| quote: | Originally posted by mrothacher
NP, Fred: I hope my desciption was accurate. I'm working on a "Poor Guy's" Audio Precision test rig |
Poor is what you are after shelling out for an AP. |
|
|
| lumanauw |
Still from the same paper, what is IM analyzer (featuring in Audio Express)? Is this a device to display harmonics?
How do Mr. Pass get this picture? I've always wanted to see harmonics (more than measuring it). Many people in this website can view 2nd, 3rd,even, odd, etc harmonic. How can I do that? |
|
|
| mrothacher |
| If you want to see individual harmoncs, an FFT analyzer is the best route. |
|
|
| lumanauw |
| Again, sorry about this. What is FFT analyzer? Can it be made DIY or have to buy it? |
|
|
| promitheus |
| You can still find some analog audio test equipment from Leader. They have a model with signal generator and distortion analyzer. I have to find the model number. |
|
|
| mrothacher |
To first familiarize yourself with the equipment and methods you may want to pick up a copy of "Audio Measurement Handbook"
 |
|
|
| promitheus |
| Do you the ISBN or other info on the book. |
|
|
| mrothacher |
I'm sorry. Yes, the book is published by Audio Precision and can be ordered from their website.
Try here:
Audio Precision Book |
|
|
| eLarson |
FFT stands for Fast Fourier Transform. It is a means of implementing Fourier transforms using Digital Signal Processing hardware.
A DIYer with experience in programming DSP chips could conceivably do it. I don't expect it would be an inexpensive project, but then I haven't looked at DSPs since using Motorola's 56001, about 12 years ago. Perhaps prices have come down on the older models?
Erik |
|
|
| banana |
We all have a set of decent DSP in our sound card.
Just dowdload RMAA for free and let the software do all the Math. Everything like Freq resp, THD, noise, IMD etc are done by a click. Of course, the measurement accuracy is limited by the performance of your soundcard. But remember, it's free.
There should be more people trying it out.
 |
|
|
| Bricolo |
About RMAA:
When you run a measurement, do you simply measure the device on the input, or measure the soundcard in "loopback" mode and substract this result from the device's measurement? |
|
|
| grataku |
| quote: | Originally posted by lumanauw
Again, sorry about this. What is FFT analyzer? Can it be made DIY or have to buy it? |
A scope displays the signal in the time domain, how the intensity change as a function of time. This is a fairly useless way to gain any visual info about the frequency content of the signal since the small harmonic components are hidden under the huge fundamental freq (your 1kHz test signal or whatever). FT is a linear transformation from the time domain to the frequency domain. Although there is an analytical formula to get the FT (fourier didn't have a computer back then), we are lazy, short on time and much prefer to use computers to get it done. The signal is digitized (the analog waveform is represented by a table of intensity of equally spaced points along the curve) and transformed into a series of peaks at various frequencies.
Ideally with no distortion one would get a single peak at the fundamental frequency. Unfortunately there are harmonic components that appear at integer multiples of the fundamental freq. luckly this are generally small 1% or so of the real signal and so you need to display the intensity of the peaks using a dB scale. (see Banana's post)
Hope that helps.
Sorry, could have sent you to the web but I felt like writing. ;) |
|
|
| banana |
| quote: | | About RMAA: When you run a measurement, do you simply measure the device on the input, or measure the soundcard in "loopback" mode and delete this result from the device's measurement? |
I was doing a "direct loopback" measurement, soundcard line-out to soundcard Line-in. So the graph is showing the performance of my cheapo on-board soundcard.
I'm still new to this RMAA stuff. As far as I've learned, it has a function to record the test tone onto a wav file or onto a CDR, so that we can perform the test using our CDP or external DAC. But I'm not sure whether the intrinsic distortion of the soundcard can be "subtracted" away or not. |
|
|
| sianturi |
| quote: | Originally posted by lumanauw
Again, sorry about this. What is FFT analyzer? Can it be made DIY or have to buy it? |
According to (Joseph??) Fourier, all 2 dimensional signals(amplitude - time) can be split into simple sinusoidal signals. FFT is basically a DFT (Discrete Fourier Transform), but with the optimized DFT algorithm to speed up the calculation process.
So, if you run an FFT on a signal, you will know the contents of frequency in the signal. A real FFT analyzer is really expensive, and I have only used it in a fully equipped lab durring my college year. You could use the simple software like banana's suggestion. Don't forget to use an appropriate low pass filter to avoid aliasing problem, and also don't forget to select an appropriate FFT windows. |
|
|
| Dale H. Cook |
| quote: | Originally posted by mrothacher
I'm working on a "Poor Guy's" Audio Precision test rig. It's a hardware / software design. |
Mike -
That is a very viable project in these days of fast processors and affordable, high-quality sound cards. I've been experimenting with a Tascam US-122 and several software packages. |
|
|
| Dale H. Cook |
| quote: | Originally posted by sianturi
According to (Joseph??) Fourier... |
Sianturi -
Essentially correct <g> - Baron Jean-Baptiste-Joseph Fourier, French mathematician, 1768-1830, known mainly for his method of analysis of periodic functions. |
|
|
| Dale H. Cook |
| quote: | Originally posted by lumanauw
Many people in this website can view 2nd, 3rd,even, odd, etc harmonic. How can I do that? |
Lumanauw -
A distortion analyzer usually has an output that can be used for further analysis of the distortion products. That output can be displayed on an oscilloscope. The image in your message appears to be that of such a display on a dual-trace oscilloscope. The top trace is the input, and the botton trace the distortion, possibly filtered through a tuneable band-pass filter so as to display only the second harmonic. The equipment I use for such a display consistes of a Hewlett-Packard 334A distortion analyzer and a Tektronix 453 oscilloscope. I do not currently own a tunable filter but am considering buying a General Radio 1232-A tuned amplifier and null detector. |
|
|
| Nelson Pass |
You can trim for equal current if you don't have a distortion
analyzer and the amp will still work, but the lowest figures
come from trimming with an analyzer.
The alternative is a distortionless inductor (read: air core) which
would not require such trimming.
:cool: |
|
|
| Dale H. Cook |
| quote: | Originally posted by Nelson Pass
The alternative is a distortionless inductor (read: air core) which
would not require such trimming. |
Nelson -
Hmmmm... I know plenty of sources for iron, but I suppose you'd have to wind your own air core, right? It's a pity that Morris coil winders have gotten so pricey. |
|
|
| Nelson Pass |
You want about 1 H of inductance for wideband audio, but
if you're only going to use it for a tweeter amp, it gets a lot
easier.
The easiest way to create this air core transformer is to buy a
big roll of speaker wire. The two wires on the cable each form
1/2 of the coupled inductor that you want, and one side of one
wire is attached to the other end of the other wire to form the
center tap.
By the way, if you use air core inductors, and even some other
transformers I've tried, you want to consider some power
resistors to ground at both outputs for stability. I've seen two
examples of common-mode oscillation depending on the
particular inductor. A good value is maybe 47 ohms or so,
rated at three times the value of the idle dissipation
(the common-mode DC output voltage squared divided by 47).
:cool: |
|
|
| carpenter |
I just purchased this unit on Ebay. Now, can someone tell me how to use it to adjust the pots on the ZVT-7?
Hoping to hear from someone soon,
John:D |
|
|
| flg |
I beleive I've used this instrument. Many moons ago. Maybe an older version. Newer versions I think got more automatic. What does it look like. Pic? How$$
The units I used had no built in oscilator. You will need to start with a quality source. You will need to feed an input (output of amp) to it and set a ref level of 0db. There is a switdh to pull in a notch filter the notch needs to out when setting ref level. Then switch it in. The notch filter is then tuned by the giant knob on the right. The meter should drop when you are tuned to the fundamental freq. You need to critically tune the notch for the lowest meter reading you can get. You are basically filtering out the oscilator freq. The measurement you acheive is the resultant distortion and noise rms voltage... It's been a long time since I've seen one of these but that's basically what I remeber...Have fun |
|
|
| carpenter |
This unit has an auto null feature. It's the top of the line for its generation (1965 or so). It's only good to 0.1 % distortion, but it's all I can afford to spend at the moment. I figure I'll turn up the amp volume to raise the distortion level to a readable point on the scale.
It only set me back $150.00.
I've been reading posts that suggest using a 1khz sinewave cd/player. It's distortion levels are supposedly quite low.
Problem is, I don't really know what I'm doing with those pots. I'll be doing a bunch of head scratching, that's for sure.
Again, thanks.
John:D |
|
|
| Alain Dupont |
Hi,
There is a Pcb at AudioXpress from Richard Crawford to make a IM Analyser
AudioXpress march 04... plus an addenda from f Glabach
I already built-it, have to finish the box....
it has 9KHz plus 19 and 20 KHz and for 40 U$PCb plus about 100U$ components
you can build a 1PPM Distortion Analyzer ; works with a analog output ...
http://www.audioxpress.com/bksprods/pcbs/testing.htm
On the other side I bought 2 old HP 332A distortion analysers last week
with the service manual for less than 200cad$ it's capable of better than 0.1%
One of them is working pretty well, the other already scrapped!
{only kept the PCb's and average components for spare parts}
regards.
Alain. |
|
|
| carpenter |
What a bunch of bs that's turned out to be--poorly packaged (one inch of wadded paper in a single-wall corrugated container), missing pieces, blackened leads on the circuit boards, scam written all over the dam_ thing. He says he'll refund my money and that I'm not to bother him or purchase anything from him again. AS IF...
Anyway, I've purchased from a different supplier a HP-332A. Does anyone have experience with this unit? I know it doesn't auto-null. What else should I know about this unit.
Hoping for the best,
John:D |
|
|
| flg |
Sorry to hear of your misfortune. I can imagine myself doing the same:( So, I've been thinking of trying the modern approach. Finally got my good sound card back on line and downloaded RMAA. Cool. I'm also trying TrueAudio's RTA(free sample level). I havent got to far yet but I guess I'll try to measure and compare few of my standalone peices vs the PC and just see... The loop back testing on the card itself seems to be working:D :D :D
Does anyone have any experience with either of these applications? Are there other free/low cost software packages to work with your PC soundcard?
The advertising seems to indicate your sound card is the weak link of the system, but they all work??? Well, my E-Mu card can hold it's own against most and the readings I'm getting just about keep up with the card specs. I would say it's looking promissing so far...
:D :D :D |
|
|
| tschrama |
Allthough I am subscribed to audioXpress, I seem to missed the AudioXpress march 04 number with a 1ppm Dist Analyzer.
Could somebody could post a scan of the schematic of that design? . You can leave out the values to save your karma ;) |
|
|
| carpenter |
I've downloaded riaa, but I'm not crazy about the idea of routing my class A amp's voltage/current through my sound card. I suppose you could feed the amp under observation a weak signal, thus a weak output, but this seems counter productive when you want to test larger outputs.
Let me know if you discover a solution.
John:D |
|
|
| Alain Dupont |
tschrama,
Here it is, a little extract...
Alain. |
|
|
| tschrama |
Great, thanks. Gonna read at immideiately.
Are the opamp types specified?
grz,
Thijs |
|
|
| Alain Dupont |
tschrama,
CI's
1 * M74HC390
1 * M74HC163
2 * M74HC393
1 * 74ACT00B
6 * NJM5532D
1 * OPA2132PA {Digikey}
1 * NJM5534D
1 * NJM7812FA
1 * NJM7805FA
1 * NJM7912FA
ALL FROM Mouser...
Alain. |
|
|
|
