Start an analyzer software project out on Github?
I believe your answer is buried back in posts #534 and #535:
http://www.diyaudio.com/forums/equi...zer-ak5397-ak5394a-ak4490-11.html#post4810922
A question came up whether the autoranger in a different project thread would be needed. Jens said the analyzer has software APIs that would allow external software to control the analyzer's attenuator. Someone would have to write some code to do that. In the meantime you could use that autoranger project:
http://www.diyaudio.com/forums/equipment-tools/299635-autoranger-soundcards.html
as a front-end add-on.
I'm seeing the usefulness here for an open-source set of community-developed software to run various things on the analyzer! 😀 Any software gurus out there want to start a software project out on Github or wherever things like this live these days? The software could control a few specific functions on the analyzer, like this input autoranging, that the general purpose audio programs like Audio Tester wouldn't have.
I believe your answer is buried back in posts #534 and #535:
http://www.diyaudio.com/forums/equi...zer-ak5397-ak5394a-ak4490-11.html#post4810922
A question came up whether the autoranger in a different project thread would be needed. Jens said the analyzer has software APIs that would allow external software to control the analyzer's attenuator. Someone would have to write some code to do that. In the meantime you could use that autoranger project:
http://www.diyaudio.com/forums/equipment-tools/299635-autoranger-soundcards.html
as a front-end add-on.
I'm seeing the usefulness here for an open-source set of community-developed software to run various things on the analyzer! 😀 Any software gurus out there want to start a software project out on Github or wherever things like this live these days? The software could control a few specific functions on the analyzer, like this input autoranging, that the general purpose audio programs like Audio Tester wouldn't have.
Last edited:
@agdr
Thanks for handling the question before I got back to the PC 🙂
@arpagon
Having also 1 dB steps would add to the complexity. And with the large dynamic range available, I think that it would only be in rare cases that you would actually see any benefit. With an FFT analysis of the (24 bit) input signal you get a lot of resolution.
And doing the calculation from dBFS to dBV would not exactly become easier with 1 dB steps.
Thanks for handling the question before I got back to the PC 🙂
@arpagon
Having also 1 dB steps would add to the complexity. And with the large dynamic range available, I think that it would only be in rare cases that you would actually see any benefit. With an FFT analysis of the (24 bit) input signal you get a lot of resolution.
And doing the calculation from dBFS to dBV would not exactly become easier with 1 dB steps.
There are alot of great software for audio measurements out there already, I guess the only thing you are asking for is a "control panel" for this specific measurement tool right ? Could be nice with a tool sitting in the toolbar and click it and it opens up a small window to adjust all driver related settings including stuff like the attenuator.. question is, can the attenuator be reached through the standard driver, or do we have to adress it with a direct memory adress ?
It can be reached via the USB HID interface.
And yes, it would be great if some of the existing audio analyzer programs would include an option to monitor/control the attenuator settings.
The RTX6001 does come with a control panel (for Windows), which can monitor/control the settings. But it would of course be a lot better if this was integrated into the analyzer program, for correct scaling and perhaps new features like autoranging.
And yes, it would be great if some of the existing audio analyzer programs would include an option to monitor/control the attenuator settings.
The RTX6001 does come with a control panel (for Windows), which can monitor/control the settings. But it would of course be a lot better if this was integrated into the analyzer program, for correct scaling and perhaps new features like autoranging.
@JensH: Yes, I know that it will add complexity. Frankly said, I don't mind the current state of the input attenuator. Almost all of my projects are preamplifiers/headphone amplifiers/power amplifiers and in most cases I can dial the input knob till get 0 dB on each range. However I have a colleague who deals mainly with DACs and he murmurs that there are conditions where tweaking the analyzer's input level is not possible.
I'm not familiar with your PCB design, routing etc. Can you tell me (hypothetically) is it easily possible a simple stepped attenuator (for example) to be added to the unit without compromising achieved noise floor, THD etc.? I hope you catch my point, don't you?
I'm not familiar with your PCB design, routing etc. Can you tell me (hypothetically) is it easily possible a simple stepped attenuator (for example) to be added to the unit without compromising achieved noise floor, THD etc.? I hope you catch my point, don't you?
It quickly grows into quite a project. Assuming that the challenges in terms of THD and noise, as you mention, can be met, it might be possible with e.g. an additional PCB between the main board and the ADC board. But the additional attenuators would also have to be controlled, so the firmware and the PC SW would need to be changed, unless it was controlled via additional wires, which would pretty soon become messy.
If it turns out that there is a pressing need for it from a large group of customers, it could perhaps be implemented in a future version. But we would need to see a business case for it, before going down that road.
If it turns out that there is a pressing need for it from a large group of customers, it could perhaps be implemented in a future version. But we would need to see a business case for it, before going down that road.
Oh, I perfectly understand you. You did a good product after all and it's not feasible to make substantial changes in the 90'th minute 🙂.
I even do not think about changes in firmware at the moment. Only manual changes in attenuation.
And I just want your estimate (since you're the designer) - is it possible, as you said, to put additional board/stepped attenuator, etc. without compromising the achieved parameters.
The warranty is 1 year....
I even do not think about changes in firmware at the moment. Only manual changes in attenuation.
And I just want your estimate (since you're the designer) - is it possible, as you said, to put additional board/stepped attenuator, etc. without compromising the achieved parameters.
The warranty is 1 year....
I think it would be hard to estimate the effects of an additional attenuation control before doing it, when board layout can effect the distortion level, an off-board arrangement can be unpredictable.
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
Hi soongsc,
Completely agree! Adding more might be like opening a can of worms when it comes to performance and accuracy.
-Chris
Completely agree! Adding more might be like opening a can of worms when it comes to performance and accuracy.
-Chris
I didn't know it comes with control panel software! Well that is great news. 🙂 Good point, autoranging would really be best as a "plug-in" to existing tester program(s).
I was looking at the switched levels, they seem quite adequate for most of my needs. But I do wonder, if software control panel is used, will the settings stay the same when you power on the next time?
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
I had this discussion already with Jens 😀 Also until today I did not get any SW API for review.
In addition, the point is this, that two gentlemen from this forum already asked for new features within my SW and finally those did not ordered. This situation change my mind as 😱 In other words, I will not add & support such features for free without any certain amount of xx in a pre ordered SW base.
Implementing some auto ranging (input & output) in conjunct with the existing GUI & API, this has to be reviewed anyway. Currently I have not seen any documents to deal with this API.
Currently I have an input calibration and free custom gain/attenuation field each store able as a preset.
Just my 2 cents on this
Hp
Is it possible to blow the input buffer when using an excessive gain setting for the input signal?
Hi PWatts,
Probably not unless you exceed the common mode range of the chips. Even then, I think Jens mentioned that there is some over-voltage protection. It has a robust input that shouldn't be damaged unless you really abuse it.
For example. If you're measuring an amplifier output, it would pay to set the sensitivity to minimum and the attenuation to maximum. Given that you should know approximately what to expect there, you could set the attenuation range beforehand. Make it part of how you operate the instrument.
Many test instruments can be damaged by voltages exceeding 5 V peak, and those are very expensive instruments. You quickly learn to pay attention to what you are doing. Just the calibration of these things exceeds $1,000. Doesn't include repairs either.
Example, I set the input attenuation of my 'scope to the expected range, or maximum if I'm not certain. Sometimes I forget if there is someone talking to me.
Hi HpW,
Sounds good to me. I understand completely where you are coming from. Much of that control would be done by whatever software package is controlling it, right? It could default to the highest range on startup to avoid damage. Manual range chaning would be okay by me.
-Chris
Probably not unless you exceed the common mode range of the chips. Even then, I think Jens mentioned that there is some over-voltage protection. It has a robust input that shouldn't be damaged unless you really abuse it.
For example. If you're measuring an amplifier output, it would pay to set the sensitivity to minimum and the attenuation to maximum. Given that you should know approximately what to expect there, you could set the attenuation range beforehand. Make it part of how you operate the instrument.
Many test instruments can be damaged by voltages exceeding 5 V peak, and those are very expensive instruments. You quickly learn to pay attention to what you are doing. Just the calibration of these things exceeds $1,000. Doesn't include repairs either.
Example, I set the input attenuation of my 'scope to the expected range, or maximum if I'm not certain. Sometimes I forget if there is someone talking to me.
Hi HpW,
Sounds good to me. I understand completely where you are coming from. Much of that control would be done by whatever software package is controlling it, right? It could default to the highest range on startup to avoid damage. Manual range chaning would be okay by me.
-Chris
The beef question would be: If 2 applications will do an auto ranging or automatic alternation of settings and there is no such control or information given to each other app, we may get soon or later a flipper box 😱
Hp
That is correct.
@HpW,
I did offer to send you the API back in August last year (and again in November), but at the time you were not really interested.
I assume that one of the features you mention, was the support for high sample rates. I think that you were the one, who was driving this. I merely suggested a way to achieve this, a kind of work-around when using the USBStreamer (using 4 I2S channels in combination to achieve a data throughput equivalent to 768 kHz sample rate). I didn't put this forward as a request. And when I decided to change from the AK5397 to the AK5394A to get a better performance, my hardware could not even support it anymore. And the hardware that could (AK4490 + AK5397) turned out, at closer inspection, to have a very poor frequency response at the high sample rates.
I would definitely like to see your SW support the range settings to get a proper scaling.
Whether it makes sense for you, in terms of business, is of course entirely up to you. If you are interested, just let me know and we can work out the details together.
Anyway, if you are interested, I can forward the latest API to you tomorrow. I don’t have access to it today.
@PWatts
As Chris writes, it has a robust input that shouldn't be damaged unless you really abuse it. I have tested it by selecting the sensitive ranges and apply > 100V rms at 50 Hz without any problems. But of course, it is always good to be careful 🙂
Jens
Given the measurement system's dynamic range 10dB is more than sufficient, 20dB steps would likely be acceptable as well. Remember the attenuators are not volume controls, they are present to allow of some optimization of available dynamic range and for range extension.
Will this device look like a sound card to windows? I use Spectra Plus FFT software with an ESI Julia at the moment. Do you expect that Spectraplus will work with this device?
Cheers...Doug
Cheers...Doug