stinius said:
Yes, but warranty would be impractical to actualize.
If possible, I prefer buy with local warranty.
Joshua_G said:
If you can stretch to around $450-$500 mark, look at the RME Hammerfall DSP 9632
http://www.rme-audio.de/en_products_hdsp_9632.php
and there is a local rep in Israel according to RME web site.
Alex
x-pro said:
Thanks, However Asus Xonar D2/DX and EMU 0404 have similar specs (THD, Noise and DR), cost much less and available locally.
Joshua_G said:
I would not compare these to RME. IMHO -if you need a tool, buy RME, if you need an entertainment , buy EMU. I have an internal 0404 here and can send it to you for free if you wish - I have no use for it.
Alex
I would agree that for measurements the RME brings a small improvement in performance. For recording music its another story. The emu 0202 is very good in basic performance. No Linux drivers that work last I looked. Windows drivers work fine however.
Hi John,
I would have to scan them to do that, and I really won't have the time for a bit and I also don't want to damage the book. It even has nice covers! 🙂
However, what John says is very true. The entire reason I search out older books when I can afford them is the increased amount of information. More than we get from the current *.pdf data sheets. The other reason is that I love reading from books and enjoy learning about the earlier state of knowledge and beliefs.
Since I am limited mostly to reading and limited movement, please don't throw out your old data books! I am building my library and will gratefully accept them. Any old semiconductor book(s) (not so much uP's and memory stuff. More analog, discrete and tube information), ARRL, service (except TV's) and application notes. HP manuals and app notes as well. Bought some but I'm currently out of $$. Even audio service manuals that can be studied. You would never believe what you can sometimes find in service manuals in the way of circuitry.
John, the other point you have brought up about the average product design engineer I regretfully have to agree with. I have been watching a very steady decline in the quality of circuit design and documentation. This is a real trend that costs the average consumer more than people want to think about. In fact, many "high end" audio designs are so flawed that it should have been criminal to release them. They were designed to fail. I know this because of the fact that I was in a support position for much of my life. The same is occurring in test and measurement and telecommunications systems. Just look at a Nortel phone system!
There is a very active core of excellent design engineers out there as well. Our signal processing performance and transistors are getting better as a result of good engineering. I am convinced that the poor quality of design stems from both a lack of understanding, or feel, for real products coupled with a poor understanding of cost control and a lack of pride in one's own work.
So when John says "No wonder engineers are so ignorant these days.", I would modify that statement to apply to certain engineers. Many do an excellent job, but then they are probably lead by a guy who is long in this world. Just wait until we lose their guidance.
The exact thing can be said for the state of repair technicians, but then we don't give them much to be proud of, now do we?
Sorry, another topic close to my heart.
-Chris
I would have to scan them to do that, and I really won't have the time for a bit and I also don't want to damage the book. It even has nice covers! 🙂
However, what John says is very true. The entire reason I search out older books when I can afford them is the increased amount of information. More than we get from the current *.pdf data sheets. The other reason is that I love reading from books and enjoy learning about the earlier state of knowledge and beliefs.
Since I am limited mostly to reading and limited movement, please don't throw out your old data books! I am building my library and will gratefully accept them. Any old semiconductor book(s) (not so much uP's and memory stuff. More analog, discrete and tube information), ARRL, service (except TV's) and application notes. HP manuals and app notes as well. Bought some but I'm currently out of $$. Even audio service manuals that can be studied. You would never believe what you can sometimes find in service manuals in the way of circuitry.
John, the other point you have brought up about the average product design engineer I regretfully have to agree with. I have been watching a very steady decline in the quality of circuit design and documentation. This is a real trend that costs the average consumer more than people want to think about. In fact, many "high end" audio designs are so flawed that it should have been criminal to release them. They were designed to fail. I know this because of the fact that I was in a support position for much of my life. The same is occurring in test and measurement and telecommunications systems. Just look at a Nortel phone system!
There is a very active core of excellent design engineers out there as well. Our signal processing performance and transistors are getting better as a result of good engineering. I am convinced that the poor quality of design stems from both a lack of understanding, or feel, for real products coupled with a poor understanding of cost control and a lack of pride in one's own work.
So when John says "No wonder engineers are so ignorant these days.", I would modify that statement to apply to certain engineers. Many do an excellent job, but then they are probably lead by a guy who is long in this world. Just wait until we lose their guidance.
The exact thing can be said for the state of repair technicians, but then we don't give them much to be proud of, now do we?
Sorry, another topic close to my heart.
-Chris
Personally, I don't care about non-linear beta anymore. I can find what I need independently. It would have been useful if a REAL example of a typical NPN transistor commonly used here was presented in a linear graph, so its characteristics would be obvious, but it doesn't help my enquiry, and it seems to be leading to a dead end.
I am fairly tolerant of amateur and professional hi fi design, and I find with few exceptions that they are at the same level. It is one of the reasons that I have found employment in this business, as many interested parties have not the background or inclination to study engineering-physics so that they can design sophisticated equipment, and those who are qualified, often show little respect for audio design, and barely lift a finger to help it, often thinking that deriding it will give people a more realistic picture. This, of course, does little to improve design quality.
I am fairly tolerant of amateur and professional hi fi design, and I find with few exceptions that they are at the same level. It is one of the reasons that I have found employment in this business, as many interested parties have not the background or inclination to study engineering-physics so that they can design sophisticated equipment, and those who are qualified, often show little respect for audio design, and barely lift a finger to help it, often thinking that deriding it will give people a more realistic picture. This, of course, does little to improve design quality.
Well, back to removing the 7th harmonic from my ST HD analysis system. NOW, I am in a position to take my ST1700B offline and work on it. I think that I will try adding the current source on the output on the AD797. I don't need no 'stinkin' PEAK BETA feedback to give it a try.
Concerning scopes.
I just ordered locally the following scope:
http://www.m-r-c.co.il/media/uploads/PDS5022SPEC.PDF
For equivalent to US$434 and with local taxes US$502 plus US$14 shipping.
It has 8 bits vertical resolution and DC accuracy of +/- 5%.
I'm looking for additional PC scope and consider 1 of the following 2:
1. http://www.tiepie.com/downloads/documents/HS4_spec.pdf
With 12 – 16 bit resolution and DC accuracy of 0.2% +/- 10 counts. It costs EURO 648 for the 5 MHz version, which is equivalent to US$840.
2. PicoScope 4224 http://www.picotech.com/picoscope4000-specifications.html
With 12 – 16 bit resolution and accuracy of +/- 1%. It costs abut $823.35.
Any opinions about which may be better choice?
I just ordered locally the following scope:
http://www.m-r-c.co.il/media/uploads/PDS5022SPEC.PDF
For equivalent to US$434 and with local taxes US$502 plus US$14 shipping.
It has 8 bits vertical resolution and DC accuracy of +/- 5%.
I'm looking for additional PC scope and consider 1 of the following 2:
1. http://www.tiepie.com/downloads/documents/HS4_spec.pdf
With 12 – 16 bit resolution and DC accuracy of 0.2% +/- 10 counts. It costs EURO 648 for the 5 MHz version, which is equivalent to US$840.
2. PicoScope 4224 http://www.picotech.com/picoscope4000-specifications.html
With 12 – 16 bit resolution and accuracy of +/- 1%. It costs abut $823.35.
Any opinions about which may be better choice?
anatech said:
Hi Chris,
Here are a few more details regarding how I do distortion measurement and fight noise. I hope this helps. Most of what I’ll talk about is measurement of THD, but I’ll also touch on 19+20 kHz CCIF. All of this is an analog-centric approach. However, the analog spectrum analyzer I use could clearly be replaced by a digital sound-card-based spectrum analyzer.
The figure below in this post shows my overall setup for THD measurement with the maximum sensitivity I can achieve. It is built around my THD analyzer, which by itself can see down to below 0.001% out to THD-20 in a 200 kHz bandwidth. It filters the residual with a tracking second-order LPF at high frequencies. This means that it will read 20 kHz THD out to the 10th harmonic.
I precede the THD analyzer with my Distortion Magnifier (DM), whose block diagram is shown in the next post. It subtracts a version of the input signal from a scaled version of the signal from the DUT to form a deep (>60dB) null at the fundamental. A controlled amount of the source signal is then added back to achieve a known amount of distortion magnification, either 20 dB or 40 dB. This also provides fundamental energy needed by the distortion analyzer to lock onto.
The DM magnifies the distortion of the DUT without magnifying the distortion (and noise) of the oscillator. My analyzer’s lowest range is 0.003% FS. If I use the DM in front with 40 dB magnification, that becomes 0.00003% FS. Of course, this is generally noise limited.
The DM includes coarse and fine amplitude and phase adjustments to optimize the null. The phase adjustment is simply a first-order LPF that constitutes a very simplified model of the amplifier DUT. Amplitude and phase must be tweaked a bit if the fundamental frequency is changed. Because of the simple model, there will be some amplitude and phase error at the higher harmonic frequencies that will detract a bit from the cancellation process, allowing a bit of oscillator noise and distortion to come through. A more complex model of the DUT high-frequency response could be used, but I have not found that necessary for my use. Similarly, the DM could incorporate low-frequency phase matching for low-frequency THD measurements on AC-coupled power amplifiers, but that has not been implemented.
I then send the distortion residual out from my THD analyzer to the HP3580A spectrum analyzer to knock out the noise and see the individual spectral components.
For THD-20, the use of the spectrum analyzer is too limited, since it only goes out to 50 kHz. The use of the DM in front of the THD analyzer still helps greatly, largely causing the noise floor for THD-20+N to be limited by DUT noise.
For evaluation of amplifier high-frequency nonlinearity, I prefer 19+20 kHz CCIF with spectrum analysis, just like what is done in the Stereophile reviews. Here I use a custom-built twin-tone generator with a very low distortion summer. This becomes the source for the DUT and the DM. The output of the DM is then fed into the spectrum analyzer. The analyzer is rated for 80 dB dynamic range. If the DM is set to 40 dB magnification, the combination ends up with a usable dynamic range of 120 dB. The DM also has a switch position where NO source signal is added back into the path, leaving only the 60+ dB nulled signal to go into the spectrum analyzer (which does not need any fundamental to lock onto).
Hope this helps,
Bob
Attachments
For my limited knowledge, the first distortion magnifier was used here: Jan Lohstroh and Matti Otala, "An audio power amplifier for ultimate quality requirements." IEEE Transactions on Audio and Electroacoustics, Vol.AU-21, No.6, December 1973, pp. 545 - 551.
Does anybody aware about earlier implementation of this idea?
Does anybody aware about earlier implementation of this idea?
Hi Joshua,
Your option #1 looks interesting at lower cost. That is normally the way you like to go.
They are both good solutions, but the software will make or break the usefulness. For that reason, the PicoScope may actually be the best choice. There is a large installed base for these too.
Most digital 'scope are 8 bit vertical resolution until you hit things like Agilent and Tek. They have a 12 bit mode.
Hi Bob,
Thank you very much for responding. I've got a copy now and I'll be able to try these things out with my equipment. You may have helped me quite a bit here!
-Chris
Edit: I've been looking at the residual output from the THD meter for years. That alone is a huge advantage over simply noting the number.
Edit #2: Bob, I have been watching your site for the pages to be completed. You didn't think anyone actually visited your site, did you?
Anyway, I just checked again to see if I could give Joshua a link and download that information for myself also. No such luck, still under construction in those areas.
Your option #1 looks interesting at lower cost. That is normally the way you like to go.
They are both good solutions, but the software will make or break the usefulness. For that reason, the PicoScope may actually be the best choice. There is a large installed base for these too.
Most digital 'scope are 8 bit vertical resolution until you hit things like Agilent and Tek. They have a 12 bit mode.
Hi Bob,
Thank you very much for responding. I've got a copy now and I'll be able to try these things out with my equipment. You may have helped me quite a bit here!
-Chris
Edit: I've been looking at the residual output from the THD meter for years. That alone is a huge advantage over simply noting the number.
Edit #2: Bob, I have been watching your site for the pages to be completed. You didn't think anyone actually visited your site, did you?
Anyway, I just checked again to see if I could give Joshua a link and download that information for myself also. No such luck, still under construction in those areas.
Bonsai said:
#4 move cables farther away from AC mains cables
#5 move cables from metal water/drai pipes
#6 move cables from metal hangers/screws/construction/rebar materials.
Thank you, Chris.
It looks like I'll go for the PicoScope.
As for lab scope, I cannot afford a Tek or Agilent one. I know you get what you pay for, however, my means are extremely tight. What I pay for measuring equipment goes off my humble savings for older days.
It looks like I'll go for the PicoScope.
As for lab scope, I cannot afford a Tek or Agilent one. I know you get what you pay for, however, my means are extremely tight. What I pay for measuring equipment goes off my humble savings for older days.
- Status
- Not open for further replies.