I think $300 for a 334 is high. You should be able to get a self tuning auto ranging unit for that- like a Tek or a KH or a ST 1700 with a generator included for that price. Unless you are collecting for collecting's sake these are just tools and you want the most tool for your money. Or a good soundcard (juli@) and software. I used RMAA to test FM transmitters which worked really well.
I’ve been spending some time reading
the posts here as it pertains to the analyzer software.
Please correct me if I’m wrong but the “overall” impression
I get is that it’s not really a full blown replacement for a good analog distortion analyzer. This is because of sound card being the weak link of the setup because of its limited input capabilities unless of course you have a buffer between the signal source and the input. I seen someone here suggest that you can take the output of an analog distortion analyzer and run it into the input of the sound card essentially using the distortion analyzer itself as the buffer rather than building one using a resistance network or op amp. Have you or others actually tried or seen this done?
Thanks
the posts here as it pertains to the analyzer software.
Please correct me if I’m wrong but the “overall” impression
I get is that it’s not really a full blown replacement for a good analog distortion analyzer. This is because of sound card being the weak link of the setup because of its limited input capabilities unless of course you have a buffer between the signal source and the input. I seen someone here suggest that you can take the output of an analog distortion analyzer and run it into the input of the sound card essentially using the distortion analyzer itself as the buffer rather than building one using a resistance network or op amp. Have you or others actually tried or seen this done?
Thanks
It all depoends on what you are doing with it. The input overload limits what you can connect to the soundcard so an input conditioner is necessary to prevent overload and scale for optimum dynamic range.
I have used a soundcard with a distortion analyzer and it works, but not for my regular use. I have a self contained FFT analyzer (Wavetek 7530A) connected to the output of my analyzer. (Boonton 1120). But your needs may well be different.
I have used a soundcard with a distortion analyzer and it works, but not for my regular use. I have a self contained FFT analyzer (Wavetek 7530A) connected to the output of my analyzer. (Boonton 1120). But your needs may well be different.
Look around for an HP339, sometimes these show up for not a whole lot more money than the 334 you mention and it is a much nicer, and more modern instrument.
I have an old Amber 3501A - this is quite a nice instrument as well, provided you buy one in good working order or can fix it. I recently overhauled mine and it is working almost as good as when new. Amber is long out of business, but there is a small but growing cadre of enthusiasts who work on these and share information.
One of the better M-audio 24 bit pci cards or EMU 24 bit cards would make a good choice for use with audiotester or arta software and are very capable.
Front end conditioning is a must with a sound card, but with a little thought in design you can get extremely good results using this approach. Personally I think this is far more useful than most simple thd analyzers. (In fact I used mine to check the old Amber as I was repairing it.)
The ultimate of course would be a 2700 series Audio Precision, but even the older System Ones are awesome. (I'd love to own one - way beyond my budget though.)
I have an old Amber 3501A - this is quite a nice instrument as well, provided you buy one in good working order or can fix it. I recently overhauled mine and it is working almost as good as when new. Amber is long out of business, but there is a small but growing cadre of enthusiasts who work on these and share information.
One of the better M-audio 24 bit pci cards or EMU 24 bit cards would make a good choice for use with audiotester or arta software and are very capable.
Front end conditioning is a must with a sound card, but with a little thought in design you can get extremely good results using this approach. Personally I think this is far more useful than most simple thd analyzers. (In fact I used mine to check the old Amber as I was repairing it.)
The ultimate of course would be a 2700 series Audio Precision, but even the older System Ones are awesome. (I'd love to own one - way beyond my budget though.)
Boonton analyzer resource
Software for the Boonton 1120/1121 Analyzer can be had here:GPIB
And I posted layouts and instructions for filter boards for them here:
PCB Schematic Assembly list
The layout is in expresspcb format (. The boards are $51 for three in 3 days.Free download ).
Software for the Boonton 1120/1121 Analyzer can be had here:GPIB
And I posted layouts and instructions for filter boards for them here:
PCB Schematic Assembly list
The layout is in expresspcb format (. The boards are $51 for three in 3 days.Free download ).
Thank you for the links,
The one for the assembly list doesn’t appear to be working.
Can it be had someplace else?
The one for the assembly list doesn’t appear to be working.
Can it be had someplace else?
I've been using an M-Audio Firewire Audiophile external box with RMAA for a bit now. I'm curious as to what kind of protection or conditioning needs to be done on the input? So far I haven't had any problems with it, but I don't want to damage the inputs. Any suggested schematics, etc. for this?
I've used it both for amplifier (i.e., preamp or headphone amp) distortion measurements as well as used it to set the biasing (distortion measurements) on my counterpoint preamp (a little slow, but it worked).
Thanks,
Chris
I've used it both for amplifier (i.e., preamp or headphone amp) distortion measurements as well as used it to set the biasing (distortion measurements) on my counterpoint preamp (a little slow, but it worked).
Thanks,
Chris
Spectrum
I am reviewing the 3581a manual....forgive my novice-ness....to "get" or "see" results from this unit (plot/graph), a digital storage scope or XY plotter is needed too ? I only have an analog scope.
-----------
I've been buying stuff lately (so I have a lot of manual reading to do)... these:
HP 3581a ($20)
Spectral Dynamics SD345 II ($600)
Spectral Dynamics SD375 II ($100)
Hrohn-Hite 3800 tracking filter ($100)
http://i5.photobucket.com/albums/y177/Midiot/DSCN2695.jpg
I think I paid too much for the $600 spectrum analyzer, because I just got it's bigger brother (SD375) for $100.....go figure ??
------------------
The SD3xx above looks like a good budget Ebay find....they pop up now and then. I made sure the buyer would guarantee it would not be DOA, serviceability might be difficult. Mine work great so far.
http://www.cognitivevision.com/legacy.htm
=RR=
Hang it on the output of the analyzer....will tell me what ?
I am reviewing the 3581a manual....forgive my novice-ness....to "get" or "see" results from this unit (plot/graph), a digital storage scope or XY plotter is needed too ? I only have an analog scope.
-----------
I've been buying stuff lately (so I have a lot of manual reading to do)... these:
HP 3581a ($20)
Spectral Dynamics SD345 II ($600)
Spectral Dynamics SD375 II ($100)
Hrohn-Hite 3800 tracking filter ($100)
http://i5.photobucket.com/albums/y177/Midiot/DSCN2695.jpg
I think I paid too much for the $600 spectrum analyzer, because I just got it's bigger brother (SD375) for $100.....go figure ??
------------------
The SD3xx above looks like a good budget Ebay find....they pop up now and then. I made sure the buyer would guarantee it would not be DOA, serviceability might be difficult. Mine work great so far.
http://www.cognitivevision.com/legacy.htm
=RR=
The audio analyzer removes the fundamental so you don't need the dynamic range and then the other stuff is easy to see.
Those FFT's have a substantial learning curve. They can do a lot but the controls are not simple. Do they have GPIB on them? They also are limited by the A to D stage which by modern sound card standards is pretty obsolete.
Those FFT's have a substantial learning curve. They can do a lot but the controls are not simple. Do they have GPIB on them? They also are limited by the A to D stage which by modern sound card standards is pretty obsolete.
1audio said:
You mean in resolution, not range, correct ?
(IEEE 488 1978, and EIA RS-232-C: serial bit.....also these have the 1/3 and 1/1 Octave option boards.)
-----
From the manual....
Resolution: 400 lines of real-time or averaged spectrum info on each analysis range. Each channel processes a 1024 point transform.
Freq, range: 21 total ranges from 1Hz full scale, to 100kHz full scale in 1-2-4-5 sequence.
Real Time Frequency: 4kHz for single foreward transform. 2kHz for dual channel transform.
Freq. Response: + 0.5 dB at filter centers over entire freq. range up to 30kHz. + 1.0 dB at filter centers from 50 kHz to 100 kHz.
Dynamic Range: 70 dB with averaging from full scale to minimum discernible signal.
Noise Floor: Below 70 dB with averaging.
A/D Converter: 12 bits
Sampling rate: 2,56 X full scale freq. range selected on internal sampling control.
Zoom: x4 --- x100 achieved with optional board.
---
=FB=
Its the 12 bit converter. State of the art when it was made and the most expensive component in the box, today you can get 8 on a single piece of silicon that sample at much higher frequencies.
However with the preconditioning of the distortion analyzer all the stuff of interest is well within 8 bits. It does have GPIB so you can write software to get it to measure and export lots of things if you want to.
However with the preconditioning of the distortion analyzer all the stuff of interest is well within 8 bits. It does have GPIB so you can write software to get it to measure and export lots of things if you want to.
You can use the 3581A without any external indicator, although an x-y chart recorder is nice. I owned a very nice one, and paid through the nose for it on eBay in 2000, sold it last year for well under half that amount. I never used it. 
Use the meter readout, record the frequency and amplitude relative to the fundamental for all of the spurs.. Measure the noise floor between the spurs -iirc the 3581A can do that too. Then enter them in excel or a comparable open source spreadsheet (open office) and graph them. Not having tried this I can't attest to the ease of the approach relative to the effort involved, but the technique would not be too different from the way WE measured the THD of tubes like the 300B originally.
Use the meter readout, record the frequency and amplitude relative to the fundamental for all of the spurs.. Measure the noise floor between the spurs -iirc the 3581A can do that too. Then enter them in excel or a comparable open source spreadsheet (open office) and graph them. Not having tried this I can't attest to the ease of the approach relative to the effort involved, but the technique would not be too different from the way WE measured the THD of tubes like the 300B originally.
kevinkr said:
Do I need to strap on the gray beard first ?
Thanks. The meter and writing down the results is kinda what I assumed, thanks for clarifying.
=FB=
redrabbit said:
Do I need to strap on the gray beard first ?
Thanks. The meter and writing down the results is kinda what I assumed, thanks for clarifying.
=FB=
300B = Happy gray beard - and I ain't that ancient... 😀
The 3581A is a nice little analyzer, what I wanted was the one with the tube.. Never got it, now I don't need it..
Note even with the 12bit analyzer you can notch out the fundamental externally, gain up the remaining distortion products and effectively window the available dynamic range to provide additional resolution. (Effectively pushing down the noise floor by the amount of the additional gain - note however that you still have a dynamic range "window" of only 70dB or so, referenced to FS - XdB.)
Any ideas for a very quiet notch filter ?
Either a pre-made product, or DIY ?
If I go looking at some older filter units, e.g. Krohn Hite, what would I look for ?
(I'm a bit filter dyslexic...can a notch be created by a high pass and low pass filter, ...but not quite overlapping ?)
I'm searching the web now, for a DIY notch filter.
I'm also building a pink/white noise filter.
(my PC is on the other side of the house...I do have noise files on there.)
----------------
I already have these:
Wavetek/Rockland Brickwall filter (yes, it does shift the phase)
http://www.tucker.com/images/images_spec/00004498.pdf
Krohn Hite 3800 tracking filter
http://www.tucker.com/images/images_spec/00002582.pdf
=RR=
Either a pre-made product, or DIY ?
If I go looking at some older filter units, e.g. Krohn Hite, what would I look for ?
(I'm a bit filter dyslexic...can a notch be created by a high pass and low pass filter, ...but not quite overlapping ?)
I'm searching the web now, for a DIY notch filter.
I'm also building a pink/white noise filter.
(my PC is on the other side of the house...I do have noise files on there.)
----------------
I already have these:
Wavetek/Rockland Brickwall filter (yes, it does shift the phase)
http://www.tucker.com/images/images_spec/00004498.pdf
Krohn Hite 3800 tracking filter
http://www.tucker.com/images/images_spec/00002582.pdf
=RR=
What do you want the notch filter for?
If you are chasing ultra low distortion you need a passive filter. There are simple twin-t filters that with fine tuning of the resistors can have very deep notches at a specific frequency. And you will spend a lot of time dialing them in. They are useful for optimizing the distortion of ultra low distortion oscillators and similar tasks but way cumbersome for regular measurements. Both Bruel & Kjaer and Daven made passive tunable filters that are OK (I have both but they are mostly curiosities).
The KH is a very interesting ultra-narrow band pass filter. It can be used to separate the harmonics in a low distortion source. It won't become a notch filter. The more conventional filters from KH can, but again not easily. Do you have a manual for it? (I have one also and no manual.)
If you are chasing ultra low distortion you need a passive filter. There are simple twin-t filters that with fine tuning of the resistors can have very deep notches at a specific frequency. And you will spend a lot of time dialing them in. They are useful for optimizing the distortion of ultra low distortion oscillators and similar tasks but way cumbersome for regular measurements. Both Bruel & Kjaer and Daven made passive tunable filters that are OK (I have both but they are mostly curiosities).
The KH is a very interesting ultra-narrow band pass filter. It can be used to separate the harmonics in a low distortion source. It won't become a notch filter. The more conventional filters from KH can, but again not easily. Do you have a manual for it? (I have one also and no manual.)
Not yet.
I am trying to find & read manuals of other tracking filters....I'm curious, even if it may not be useful in this application.
I'll try hooking things up, for a coupl'a days.
I just ran a balanced cable from my soundcard across the house, to get some noise files....to "see" what is being filtered.
I'm just guessing that will work. Trying to "see" what the filter is doing using just a single frequency is ....frustrating.
Thanks for all your help so far.
=RR=
Notch Filter Extends Spectrum-Analyzer Range
Bruce Tibbetts, Credence Systems -- Test & Measurement World, 1/1/2000
http://www.tmworld.com/article/CA187356.html
=RR=
Bruce Tibbetts, Credence Systems -- Test & Measurement World, 1/1/2000
http://www.tmworld.com/article/CA187356.html
=RR=
Tektronics TDS3032B
Can the TDS3032B be used to do the FFT analysis of the distortion measurements? How would it be best utilized?
If you had a choice between the TPS2014 and the TDS3032B for general audio work, which one would you choose? I believe the TDS3032B is the "better" unit, but just wanted other's opinions.
Is there any need for the isolated inputs of the TPS2014?
Thanks for your time,
Donovan
Can the TDS3032B be used to do the FFT analysis of the distortion measurements? How would it be best utilized?
If you had a choice between the TPS2014 and the TDS3032B for general audio work, which one would you choose? I believe the TDS3032B is the "better" unit, but just wanted other's opinions.
Is there any need for the isolated inputs of the TPS2014?
Thanks for your time,
Donovan
- Status
- Not open for further replies.