Jens, how is it progressing? Soon the somewhat crippeled QA405... Oh. Sorry QA401 hits our ePay fingers so a choice have to be made by us like me that not lightly are able to go for both... I rather go for yours but that would require some more action from your side.
Now is the time - should we go for your design or would you recommend us to put our hard earned money on Qa401? A light answer to this will follow you is my sincere guess.
Regards
Now is the time - should we go for your design or would you recommend us to put our hard earned money on Qa401? A light answer to this will follow you is my sincere guess.
Regards
Sorry, to late to add to the former post. Just keep options open for connecting external devices be they differential or single ended - your choice 
There is not more time left unless you are down to -140db'iesh. -130dB'iesh you need to show that your design is much more flexible and versatile.
Regards
There is not more time left unless you are down to -140db'iesh. -130dB'iesh you need to show that your design is much more flexible and versatile.
Regards
Last edited:
As I read the responses from Jens he has a role of designer, producer and vendor. There will be no cheap diy alternative as of the responses. Jens is actively seeking a vay to market his design and by that monetizing from it. Sorry for being harsh... So no diy more than hackning a given (bought) product. Am I right about my conclusions?
Regards
Regards
Thanks for your interest.
I am still trying to get an idea of what it will cost to manufacture the analyzer. And I think that I am pretty close to having a first estimate.
At the moment I don't rule out any possibility. I am trying to get an estimate of what a complete product will cost. As part of that I will of course get an estimate of what the mounted PCB's will cost. In principle the mounted PCB's could be delivered without a box etc. But I am not sure that this would make sense. And bare PCB's really wouldn't make sense. There are more than 800 components and most of them are surface mount. And a lot of special components, which wouldn't be cost effective to buy in small quantities.
I hope that it will be possible to offer the complete product at an affordable price, also for DIY use. I hope that I can get a ball park figure worked out within the next couple of weeks.
One of the problems is that it is difficult to estimate the quantity. Which will influence the cost. And of course the cost will influence the quantity.
I have made some minor design changes lately and I think that it is now very close to a final layout.
I have also started looking into the possibility of using the AK4497 as an alternative to AK4490 for the DAC part. And I plan to look into the possibility of using the new ESS DAC's. I don't expect the ESS based design to be ready in the near future though. First of all I need to get access to the data sheets.
So far I have only tested the analyzer on Windows. It should of course be tested on Linux and MAC at some point in time. I don't expect that it will be a problem, but of course you never know until it has been tested.
Any good suggestions for analyzer SW for Linux and MAC?
I am still trying to get an idea of what it will cost to manufacture the analyzer. And I think that I am pretty close to having a first estimate.
At the moment I don't rule out any possibility. I am trying to get an estimate of what a complete product will cost. As part of that I will of course get an estimate of what the mounted PCB's will cost. In principle the mounted PCB's could be delivered without a box etc. But I am not sure that this would make sense. And bare PCB's really wouldn't make sense. There are more than 800 components and most of them are surface mount. And a lot of special components, which wouldn't be cost effective to buy in small quantities.
I hope that it will be possible to offer the complete product at an affordable price, also for DIY use. I hope that I can get a ball park figure worked out within the next couple of weeks.
One of the problems is that it is difficult to estimate the quantity. Which will influence the cost. And of course the cost will influence the quantity.
I have made some minor design changes lately and I think that it is now very close to a final layout.
I have also started looking into the possibility of using the AK4497 as an alternative to AK4490 for the DAC part. And I plan to look into the possibility of using the new ESS DAC's. I don't expect the ESS based design to be ready in the near future though. First of all I need to get access to the data sheets.
So far I have only tested the analyzer on Windows. It should of course be tested on Linux and MAC at some point in time. I don't expect that it will be a problem, but of course you never know until it has been tested.
Any good suggestions for analyzer SW for Linux and MAC?
Hi JensH,
If you are pushing the boundaries with this, you might have to do a custom front end to unlock the full potential of it. With any luck, it might be a matter of compiling the sw for Linux (one should do all flavors), maybe the same for Mac. One thing for sure, no hidden stuff in Linux. Microsoft couldn't bear the visibility of open source. I don't think they could exist at their price points in any open environment. Translation: It might be easier for you to program for Linux and Mac.
Give us a holler if you need beta testers.
-Chris
If you are pushing the boundaries with this, you might have to do a custom front end to unlock the full potential of it. With any luck, it might be a matter of compiling the sw for Linux (one should do all flavors), maybe the same for Mac. One thing for sure, no hidden stuff in Linux. Microsoft couldn't bear the visibility of open source. I don't think they could exist at their price points in any open environment. Translation: It might be easier for you to program for Linux and Mac.
Give us a holler if you need beta testers.
-Chris
Hi Chris,
Thanks for the offer.
I don't plan to develop SW for it. The analyzer will work as a sound card with ASIO interface, which requires a driver to work under Windows. This driver will of course be provided.
I don't plan to include the analyzer SW as part of the delivery. This SW could be any available SW compatible with an ASIO sound card. I have mainly used audioTester, but I have also used HP Works and Arta. So the user should be able to choose his favorite analyzer SW or different SW depending on the measurements needed.
Thanks for the offer.
I don't plan to develop SW for it. The analyzer will work as a sound card with ASIO interface, which requires a driver to work under Windows. This driver will of course be provided.
I don't plan to include the analyzer SW as part of the delivery. This SW could be any available SW compatible with an ASIO sound card. I have mainly used audioTester, but I have also used HP Works and Arta. So the user should be able to choose his favorite analyzer SW or different SW depending on the measurements needed.
I totally agree. I am sure the benefit to the diyAudio community (and other audio designers / engineers / etc.) would be much higher if you'd provide complete working boards or even fully built boxes.
May I suggest MATAA (yes, I developed that myself). It uses PortAudio and therefore works on Mac OS X (using CoreAudio), Linux (using ALSA) and on Windows (using ASIO). MATAA is basically an extension to Matlab / GNU Octave and may therefore be a bit tricky to set up for Matlab / Octave newbies. But hey, it's free and way more flexible than all other audio testing software I have seen. According to earlier posts, your design works like a standard USB audio device, so it should work with Mac OS X and modern Linux distributions out of the box (no need for extra drivers and stuff). If you have a unit available, I'd be more than happy to give it a try on a Mac and a Linux box.
I think comparing a QA401 and Jens' project is misleading. Jen's project is capable of SOTA distortion performance and really a soundcard component for a measurement system. The QA401 is a much more complete solution but gives up a little for distortion performance. Each has its uses. Ultimately neither are complete but if you need to get on the air and making measurements the QA400 or QA401 would be a preferred choice. The QA401 comes with a system controlled attenuator that automatically scales the displayed levels teducing another point of potential confusion. it doees have full isolation between the host PC and the measurement electronics and the software does work. If he can get ASIO working then it gets extended with other software which is good.
Jen's project is very high performance but you need software to use it. That means a more complicated effort to get a measurement system running. If you are willing to undertake that effort more options open up but it won't be a plug and play experience.
Jen's project is very high performance but you need software to use it. That means a more complicated effort to get a measurement system running. If you are willing to undertake that effort more options open up but it won't be a plug and play experience.
I don't see the point here. Both the QA and Jens' design need some kind of software to work, and it will be up to the user to install and configure this on his/her computer. The way it looks to me, both the QA401 and Jens' design will work as "standard USB audio devices", so they will be recognized as such by the operating system (ok, Windows may need an extra driver for that). The user therefore has the choice to use the QA401 and Jens' design with just about any audio testing software around.
If the QA unit will turn out not to work as a "standard USB audio device", then it will likely only work with its own, vendor-specific software. Not good (and useless to me).
Hi Demian,
I agree that having system controlled attenuator settings can avoid confusion and the need for additional comments to test results etc. It is something, which I have thought of implementing and perhaps I actually should do it. It probably won't be a big effort.
The downside is that it requires a tighter integration with the analyzer SW to work. In reality this means that only some analyzer SW (if any) will support it. But I could implement a system, which gives three options:
1. Control by mechanical switches, no SW integration.
2. Control by mechanical switches, but settings readable from the analyzer SW.
3. SW control, mechanical switches disabled.
Option 2 and 3 will allow automatic scaling of the measured levels. And it would make it possible to select one or the other based on personal preference (subject to implementation in the analyzer SW available).
The control interface may require a separate USB port, to avoid potential problems when trying to transfer the control via the audio USB connection. If a USB to I2C interface is used it should be fairly straight forward to read out or control the relays. This could also be used for production test equipment, where a PC could set up the levels from a control program.
I agree that having system controlled attenuator settings can avoid confusion and the need for additional comments to test results etc. It is something, which I have thought of implementing and perhaps I actually should do it. It probably won't be a big effort.
The downside is that it requires a tighter integration with the analyzer SW to work. In reality this means that only some analyzer SW (if any) will support it. But I could implement a system, which gives three options:
1. Control by mechanical switches, no SW integration.
2. Control by mechanical switches, but settings readable from the analyzer SW.
3. SW control, mechanical switches disabled.
Option 2 and 3 will allow automatic scaling of the measured levels. And it would make it possible to select one or the other based on personal preference (subject to implementation in the analyzer SW available).
The control interface may require a separate USB port, to avoid potential problems when trying to transfer the control via the audio USB connection. If a USB to I2C interface is used it should be fairly straight forward to read out or control the relays. This could also be used for production test equipment, where a PC could set up the levels from a control program.
Hi JensH,
Either options 2 or 3 would work fine. Manual control that is readable might be the best option as you don't need to sense the level and change ranges in software. That and the machine state has "memory" so that you can continue at the same level as the last session if you wanted. I haven't been spoiled by auto-ranging audio analyzers yet. It would also prevent gain range changes during a sweep, which is annoying.
-Chris
Either options 2 or 3 would work fine. Manual control that is readable might be the best option as you don't need to sense the level and change ranges in software. That and the machine state has "memory" so that you can continue at the same level as the last session if you wanted. I haven't been spoiled by auto-ranging audio analyzers yet. It would also prevent gain range changes during a sweep, which is annoying.
-Chris
mbrennwa-
The QA400 and QA401 come with dedicated software that addresses most testing. The overall model is a modest, stripped down, APx. it removes a host of questions like "what is the voltage I'm measuring?" I have used the QA400 in several professional applications and got the answers I needed quickly, in an ultra-moble setting. I can tell the QA401 is an improvement that addresses my biggest issues with the QA400.
Jens-
Your suggestion about the input attenuator sparked an idea- what would be pretty ideal would be a separate module that can handle the input metering and attenuation using an autoranging approach. If it had more steps than the typical 20 dB steps (maybe 10 dB or even 6 dB) it would be possible to keep the input to the ADC in its optimum range and still handle a large input dynamic range. If input metering was part of this external module and could be read via USB separately then we would not need to be so concerned about the integration. A display on the module would be enough but adding the USB would allow the software to add that info to the measurement data for scaling in the resulting display. I think this may be expensive to implement since its the bulk of a traditional audio analyzer which is where the cost lies.
In any case I would push for relay ranging and usb + local control if not autoranging. Max input 100V, min range 100 mV, full differential in. Something similar for the output with 7V max (maybe 3.5V into 50 Ohms?)
The QA400 and QA401 come with dedicated software that addresses most testing. The overall model is a modest, stripped down, APx. it removes a host of questions like "what is the voltage I'm measuring?" I have used the QA400 in several professional applications and got the answers I needed quickly, in an ultra-moble setting. I can tell the QA401 is an improvement that addresses my biggest issues with the QA400.
Jens-
Your suggestion about the input attenuator sparked an idea- what would be pretty ideal would be a separate module that can handle the input metering and attenuation using an autoranging approach. If it had more steps than the typical 20 dB steps (maybe 10 dB or even 6 dB) it would be possible to keep the input to the ADC in its optimum range and still handle a large input dynamic range. If input metering was part of this external module and could be read via USB separately then we would not need to be so concerned about the integration. A display on the module would be enough but adding the USB would allow the software to add that info to the measurement data for scaling in the resulting display. I think this may be expensive to implement since its the bulk of a traditional audio analyzer which is where the cost lies.
In any case I would push for relay ranging and usb + local control if not autoranging. Max input 100V, min range 100 mV, full differential in. Something similar for the output with 7V max (maybe 3.5V into 50 Ohms?)
...and...
I agree that an attenuator would be very neat. However, no software will support that out of the box. So if you decided to add some kind of software interface, it should by all means be optional. The analyzer must be fully functional without the extra software-controlled attenuation. And the software interface should be easy to implement on all major operating systems (Windows, Linux, Macs)! I am thinking if RS232 via USB.
You might also consider to add the attenuator part as a separate module to the existing design. That might allow you to come up with a finished analyzer board/unit soon, and add the software attenuator thingy later.
Yes. But many would prefer using other software packages (ARTA, Sound Easy, RME, Audacity, RMAA, FuzzMeasure, MATAA, and many more). So I asked QuantAsylum, and they responded with this:
"The QA400 can only work with QA software. But the later devices will have large enough buffers that they can work as ASIO devices in the future on PCs. The date for that to happen isn't clear at this point."
So the QA401 will likely be usable with other software packages as well.
If I implement this it was my intention to make it optional. So it should support all 3 options. If the SW does not support attenuator control, option 1 would be the default control method. If the SW does support it, either option 2 or 3 could be used. Perhaps even user selectable if the SW supports both options.
A simple implementation could be based on e.g. a TCA9555 I2C I/O expander or similar. And the control interface could be open information allowing any SW developer to write a driver for it.