SVO @ 20Hz
This is the lamp based mult. SVO at 20Hz running just under 1Vrms no notch just the oscillator.
No AGC here just two 9V batteries and a pot to set the Vc on the multiplier.
This look okay to me.
Bob C. can you have a look and see if there any sign of thermal modulation?
That's not a hunk of third H there just power supply hum.
Cheers,
This is the lamp based mult. SVO at 20Hz running just under 1Vrms no notch just the oscillator.
No AGC here just two 9V batteries and a pot to set the Vc on the multiplier.
This look okay to me.
Bob C. can you have a look and see if there any sign of thermal modulation?
That's not a hunk of third H there just power supply hum.
Cheers,
Attachments
Last edited:
725D-
The signal flow is left to right (Counter-intuitive for an Asian design). The big caps are input coupling caps for the differential input. the two relays would give possibly 4 attenuation steps, 100,30,10,3. Note the trim caps for HF response. The critical resistors are TKD, which are very high quality resistors.
The amp circuit has diodes and lightbulbs on its input for protection, similar to the Amber circuit (among others). The input diff pair is probably in the metal can. I'm guessing but I think the input device has a series 100 Ohms resistor (I can't really read the color code in the picture). I would bet that its a differential JFET cascoded for the input. If the resistor values are what I suspect in the attenuator that is the first limitation.
The 4 relays in the upper right are the gain ranging and the module is the summing amp for the differential signal. The 5334 may be used to reduce the common mode signal.
I would go no further without an extender board. It helps a lot to see the signals.
I would bet that the board has both +/-15V and +/-30V supplies.
Except for attenuator impedance probably little opportunity to improve noise floor here.
Mine is different, it doesn't have the differential input and the relays are bigger metal cased things.
Since this may be the lowest distortion amplifier known there may be tricks to learn, but there are no magic parts or really novel circuits, just very careful execution.
The signal flow is left to right (Counter-intuitive for an Asian design). The big caps are input coupling caps for the differential input. the two relays would give possibly 4 attenuation steps, 100,30,10,3. Note the trim caps for HF response. The critical resistors are TKD, which are very high quality resistors.
The amp circuit has diodes and lightbulbs on its input for protection, similar to the Amber circuit (among others). The input diff pair is probably in the metal can. I'm guessing but I think the input device has a series 100 Ohms resistor (I can't really read the color code in the picture). I would bet that its a differential JFET cascoded for the input. If the resistor values are what I suspect in the attenuator that is the first limitation.
The 4 relays in the upper right are the gain ranging and the module is the summing amp for the differential signal. The 5334 may be used to reduce the common mode signal.
I would go no further without an extender board. It helps a lot to see the signals.
I would bet that the board has both +/-15V and +/-30V supplies.
Except for attenuator impedance probably little opportunity to improve noise floor here.
Mine is different, it doesn't have the differential input and the relays are bigger metal cased things.
Since this may be the lowest distortion amplifier known there may be tricks to learn, but there are no magic parts or really novel circuits, just very careful execution.
FYI update --- three spot freqs from the ShibaSoku AG15 generator -->
FREQ THD+N
100Hz .00005%
1000Hz .00004%
10KHz .000125%
The KH4402B with LT1037 for U141 and AD797 for osc U125 Opamp changes -->
100Hz .00007%
1000Hz .000045%
10KHz .00011%
Close but no cigar -- with OPA134 for U141, the 1KHz thd+n could be tuned down to .000025% or about -130dB -- 10 dB shy of goal but thd+n at 10KHz was .0005%
The wide R changes when selecting different osc freqs hurts thd+n figures.... just to keep C values low and cheaper. might be better to use larger values of C than large variations of values for R. [ yes, i know the problem with that but to get lowest thd+n requires lower R values].
The ad725d analyzer. The noise I will attempt is after the notch to see if the S/N can be improved at the monitor output port so more of the harmonics show up unmasked by noise. There is only about a 10db range for improvement. But 10dB is 10dB any way I can get it.
The input uses dozens of discrete transistors. No interest there except what the circuit design for the unbal and bal input amps are.... looks pretty impressive bit of analog design work going on. Probably done by a National Treasure with 10th degree Masters helping with it.
Thx-RNMarsh
FREQ THD+N
100Hz .00005%
1000Hz .00004%
10KHz .000125%
The KH4402B with LT1037 for U141 and AD797 for osc U125 Opamp changes -->
100Hz .00007%
1000Hz .000045%
10KHz .00011%
Close but no cigar -- with OPA134 for U141, the 1KHz thd+n could be tuned down to .000025% or about -130dB -- 10 dB shy of goal but thd+n at 10KHz was .0005%
The wide R changes when selecting different osc freqs hurts thd+n figures.... just to keep C values low and cheaper. might be better to use larger values of C than large variations of values for R. [ yes, i know the problem with that but to get lowest thd+n requires lower R values].
The ad725d analyzer. The noise I will attempt is after the notch to see if the S/N can be improved at the monitor output port so more of the harmonics show up unmasked by noise. There is only about a 10db range for improvement. But 10dB is 10dB any way I can get it.
The input uses dozens of discrete transistors. No interest there except what the circuit design for the unbal and bal input amps are.... looks pretty impressive bit of analog design work going on. Probably done by a National Treasure with 10th degree Masters helping with it.
Thx-RNMarsh
Last edited:
If you want to look at the harmonics the analyzer function is the answer. It can see way down into the noise to see harmonics. I think its quite relevant since our hearing has a similar ability to decode repetitive patterns amid chaotic noise. If you want to know what the noise floor of the generator is then you need to look at the monitor output.
When looking at amplifiers, in reality, the noise floor of the test system needs to only be more than 3 dB below the noise floor of the DUT. More than that won't give you any insight. If the AG16 or the 590AR noise floor (10V to residual wide band noise) is around 140 dB (what I measured I think) there are very few devices that would exceed that range that I can think of.
When looking at amplifiers, in reality, the noise floor of the test system needs to only be more than 3 dB below the noise floor of the DUT. More than that won't give you any insight. If the AG16 or the 590AR noise floor (10V to residual wide band noise) is around 140 dB (what I measured I think) there are very few devices that would exceed that range that I can think of.
I am starting with THD+N and working on getting it down and THEN go to the FFT. I can easily see the analyzer output monitor with a scope. BUt I prefer something else that can be interfaced (USB) to a computer instead. And without a scope as a go-between. Thus, waiting for you and davada to have a working ADC/FFT system for us. Meanwhile, I keep tinkering on the oscillators as a test source for other DIY'ers to have -- the ShibaSoku generator isnt available to most DIY'ers (and a new one cost $9.2K !!).... an alternative(s). And, it is possible today to get similar low THD (+N) with the KH4402B with some simple changes.
Now I also want to have a really good small/portable FFT/ADC/PC.
I have a notch filter for use with a DUT made up now.
So. Which model and brand ADC is to be used and modified? Reading the messages here, it isnt clear. Is there one or are there two models to use which has the least headache with drivers?
Thx-RNMarsh
Now I also want to have a really good small/portable FFT/ADC/PC.
I have a notch filter for use with a DUT made up now.
So. Which model and brand ADC is to be used and modified? Reading the messages here, it isnt clear. Is there one or are there two models to use which has the least headache with drivers?
Thx-RNMarsh
Last edited:
I'm getting great results from the QA400; http://www.diyaudio.com/forums/equi...n-audio-range-oscillator-201.html#post3433762 Have you been back to QA to get yours sorted out?
Which OS are you using?
For USB an EMU 0404 or 0204 would be the first choice I think. They are known commodities. While I can get results from them the interfacing is more trouble than it should be.
If you are using a desktop get an ESI Juli@. It works well, the drivers have not been an issue and I can do all manner of things with it and Praxis. It also works fine with ARTA, HPWorks and everything else I have tried.
You will need an interface like the one I am tweaking. The common ground stuff will cause no end of problems until you can disconnect the grounds.
Which OS are you using?
For USB an EMU 0404 or 0204 would be the first choice I think. They are known commodities. While I can get results from them the interfacing is more trouble than it should be.
If you are using a desktop get an ESI Juli@. It works well, the drivers have not been an issue and I can do all manner of things with it and Praxis. It also works fine with ARTA, HPWorks and everything else I have tried.
You will need an interface like the one I am tweaking. The common ground stuff will cause no end of problems until you can disconnect the grounds.
My 2cents on ADCs -- the EMU 0204 and QA400 (I have both) are useless for looking at harmonics above 40kHz and are compromised at 30kHz due to the shape of the noise floor. I had an EMU 1212M which was good and flat out to 100kHz, but the drivers and SW were really a pain, really. I sent that to David -- maybe he can make it work right.
But if I understand what you two are doing with the ShibaSoku units, then the full resolution of the ADC is not so important for the output signals you are looking at; if that's true, then I would use the EMU 0204 because it is not tied to any particular software. I like the QA400 a lot, but I also like using ARTA a lot, too.
But if I understand what you two are doing with the ShibaSoku units, then the full resolution of the ADC is not so important for the output signals you are looking at; if that's true, then I would use the EMU 0204 because it is not tied to any particular software. I like the QA400 a lot, but I also like using ARTA a lot, too.
Hi Demian,
I bought an emu 0204 first, then a qa400. the drivers are bizzar on the emu so i dropped it. Then dadava mentioned he was using the qa400 as a tester. it worked... sort of... more hw driver issues again. I have two portables one with xp (old and slow... same one I had when we went to Taiwan trips). Other portable with Win7 and one all-in-one desk model - also win7 (pro). Then the shibasoku dropped in my lap. But I still want a good portable, small light weight system. PnP that works and I wont have to fear a win update will kill my FFt system.
I am ready to try again to remove and redownload the latest and greatest sw for the Qa400.
Thx for info and clarifying what to use. -Richard
I bought an emu 0204 first, then a qa400. the drivers are bizzar on the emu so i dropped it. Then dadava mentioned he was using the qa400 as a tester. it worked... sort of... more hw driver issues again. I have two portables one with xp (old and slow... same one I had when we went to Taiwan trips). Other portable with Win7 and one all-in-one desk model - also win7 (pro). Then the shibasoku dropped in my lap. But I still want a good portable, small light weight system. PnP that works and I wont have to fear a win update will kill my FFt system.
I am ready to try again to remove and redownload the latest and greatest sw for the Qa400.
Thx for info and clarifying what to use. -Richard
Last edited:
Hi Dick,
I would use it with the analyzer hardware but also as a stand alone system... probably as my first look. I guess I can live with the BW limitations as it will not be my primary system anymore.
Your experience with the software is similar to mine and others. I got arta also and it looked great - just need a really good adc that is PnP ready. Guess i'll try to spend more time with both emu and qa models again... .
I shall return.
Thx-Richard
-Richard
aka Dick marsh
Last edited:
Richard, given that you've had driver issues with both the 0204 and the QA400, I wonder if something else is going on in your target machine? I did have driver issues early on with the QA400, but that got straightened out after some fooling around and with new software.
I still have to boot the computer and then connect and disconnect the USB plug a couple of times to get a stable connection, and I'm running XP, which should be solid now, so I can sympathize with your frustration. It is not quite PnP for me at this point, but I'm also at least one SW rev behind, now. The 0204 never gave me driver problems with XP, but maybe I was lucky -- but Creative has never made it easy to install their drivers.
I still have to boot the computer and then connect and disconnect the USB plug a couple of times to get a stable connection, and I'm running XP, which should be solid now, so I can sympathize with your frustration. It is not quite PnP for me at this point, but I'm also at least one SW rev behind, now. The 0204 never gave me driver problems with XP, but maybe I was lucky -- but Creative has never made it easy to install their drivers.
I have a DG8SAQ network analyzer that mixes down to audio. Works really
well but I usually have quite some work to make my audio system function for
audio again after use.
I bought a Terratec DMX6FIRE USB as a dedicated sound system for
experiments, but had driver problems I could not resolve
So, currently I use a SNA-33 spectrum analyzer that has a 1 Hz filter
and works from 20 Hz to 26 GHz, but the sweeps are slooooow.
I was willing to buy a FFT analyzer. There seems to be no progress in
the last 20 years. Still 16 bit/100KHZ and if you want 2 channels,
BW is often reduced to 50 KHz. The useable ones still cost half
of the new price 2nd/3rd hand.
I want one that can do cross correlation and then averaging since that
brings a lot of dynamic range at the expense of time. It is always
stunning to see the noise floor drop on a E5052B.
Is there anything comparable to a SR785?
As an interim solution, I was interested into a QA400, but,
when I read here that it has driver issues, too, it is much
less appealing. One of its interesting points was that it does
not register as a sound device. :-(
cheers, Gerhard
well but I usually have quite some work to make my audio system function for
audio again after use.
I bought a Terratec DMX6FIRE USB as a dedicated sound system for
experiments, but had driver problems I could not resolve
So, currently I use a SNA-33 spectrum analyzer that has a 1 Hz filter
and works from 20 Hz to 26 GHz, but the sweeps are slooooow.
I was willing to buy a FFT analyzer. There seems to be no progress in
the last 20 years. Still 16 bit/100KHZ and if you want 2 channels,
BW is often reduced to 50 KHz. The useable ones still cost half
of the new price 2nd/3rd hand.
I want one that can do cross correlation and then averaging since that
brings a lot of dynamic range at the expense of time. It is always
stunning to see the noise floor drop on a E5052B.
Is there anything comparable to a SR785?
As an interim solution, I was interested into a QA400, but,
when I read here that it has driver issues, too, it is much
less appealing. One of its interesting points was that it does
not register as a sound device. :-(
cheers, Gerhard
Last edited:
Thanks for this data, Richard. That ShibaSoku is a remarkable piece of equipment! How much does it go for? Did you get it used?
Cheers,
Bob
I dont know how much a new one costs.... but if their generator cost more than $9000 USD, you can imagine. I got it about 2.3542238721 micro-seconds after it showed up in eBay.
I called the only usa distributor -- who was hard to find... numbers changed... sounded like he rarely, if ever sold any ShibaSoku. he said he would find out about the availability of the generator for me. (?) A couple days later, said he 'could sell me one. ? They are built like a good piece of HP gear. Solid. I never would have known they existed if Demian hadnt got hold of an earlier model some time ago.
Thx-RNMarsh
I called the only usa distributor -- who was hard to find... numbers changed... sounded like he rarely, if ever sold any ShibaSoku. he said he would find out about the availability of the generator for me. (?) A couple days later, said he 'could sell me one. ? They are built like a good piece of HP gear. Solid. I never would have known they existed if Demian hadnt got hold of an earlier model some time ago.
Thx-RNMarsh
Last edited:
I have a DG8SAQ network analyzer that mixes down to audio. Works really
well but I usually have quite some work to make my audio system function for
audio again after use.
I bought a Terratec DMX6FIRE USB as a dedicated sound system for
experiments, but had driver problems I could not resolve
So, currently I use a SNA-33 spectrum analyzer that has a 1 Hz filter
and works from 20 Hz to 26 GHz, but the sweeps are slooooow.
I was willing to buy a FFT analyzer. There seems to be no progress in
the last 20 years. Still 16 bit/100KHZ and if you want 2 channels,
BW is often reduced to 50 KHz. The useable ones still cost half
of the new price 2nd/3rd hand.
I want one that can do cross correlation and then averaging since that
brings a lot of dynamic range at the expense of time. It is always
stunning to see the noise floor drop on a E5052B.
Is there anything comparable to a SR785?
As an interim solution, I was interested into a QA400, but,
when I read here that it has driver issues, too, it is much
less appealing. One of its interesting points was that it does
not register as a sound device. :-(
cheers, Gerhard
I am beginning to think the way to go might be a high speed transient waveform capture device... sample & hold... import it into a PC running DADisP or similar and process the waveform(s) for what ever you like.
Might not be able to do a lot of averaging with it... thus the thd+n requirment I have for myself.
But, enough averaging for most acoustics work would be easy to accomplish.
Last edited:
The Panasonics are really excellent products. I have no experience with them but they should work well for amplifier tests and be reliable and the specs are very good on the best model.Might have potential for upgardes too. A lot less expensive but not a give-away, either .See thier home page for newest models and specs. Do they have FFT's?
Last edited:
This one may look familiar to you Dick. Panasonic VP-7725B | Used Audio / Distortion Analyzer | Used Test Equipment | Refurbished Test Equipment
The similarity in features and model numbers seems to indicate they are Shibasoku!
The similarity in features and model numbers seems to indicate they are Shibasoku!
The Panasonic/Leaver stuff is interesting and borrows (licenses?) the ShibaSoku technology. ShibaSoku is a major vendor of video and broadcast test equipment. The audio stuff was a side trip for them in the late '80s it seems. They have a patent on the distortion analyzer part of the product. The products are not the same but similar. The 725 I have (first gen) is very similar internally to Richard's in circuits but the redesigned the unit with newer relays etc. along with newer features like GPIB control and two balanced/ unbalanced inputs.
The Panasonic worth making an effort for is the 7725B but its as rare as the 725 is in the west. Most of this stuff was used in Japanese engineering and manufacturing along with some in China (where the Panasonic stuff went to become Leaver).
The ShibaSoku box I'm after and will probably never find is this: http://www.shibasoku.com/download/avc/am51c_e.pdf It has the ultra generator and the analyzer in the box along with other features that make it more useful as a system. I don't have enough shop space for it so I'm not looking hard (and I have a budget limit of $300 for any of this stuff).
Meanwhile, I'm getting fine results and no driver problems with the QA400. I'm using with Win 7 /64 on a laptop. Since it doesn't use the normal audio interface it doesn't get "borked" by other audio processes in the PC. You need to run the latest firmware and install the current drivers. And you will need some form of interface unless you are working with mobile products.
The Panasonic worth making an effort for is the 7725B but its as rare as the 725 is in the west. Most of this stuff was used in Japanese engineering and manufacturing along with some in China (where the Panasonic stuff went to become Leaver).
The ShibaSoku box I'm after and will probably never find is this: http://www.shibasoku.com/download/avc/am51c_e.pdf It has the ultra generator and the analyzer in the box along with other features that make it more useful as a system. I don't have enough shop space for it so I'm not looking hard (and I have a budget limit of $300 for any of this stuff).
Meanwhile, I'm getting fine results and no driver problems with the QA400. I'm using with Win 7 /64 on a laptop. Since it doesn't use the normal audio interface it doesn't get "borked" by other audio processes in the PC. You need to run the latest firmware and install the current drivers. And you will need some form of interface unless you are working with mobile products.