5 years ago they stopped selling this tea is in any shop in Germany
But I found it again in Switzerland.
If you intend to establish this tin box as the corporate diyaudio casing
- here is a vendor who could serve reasonable quantities.New Design Lipton Tea Tin Can - Buy Lipton Tea Tin Can,Lipton Tea Tin Can,Lipton Tea Tin Can Product on Alibaba.com
Last edited:
choco, perhaps run the same test(s) with the bandwidth up at 192, and avg = 0 and avg = ~10?
It looks very free of LF pick up, and noise in general, surprisingly so.
Also, perhaps run a screen of the unit "native" in loop back, and with the input(s) shorted?
I found the two channels are slightly different, I think everyone else has as well...
Looks interesting.
_-_-
It looks very free of LF pick up, and noise in general, surprisingly so.
Also, perhaps run a screen of the unit "native" in loop back, and with the input(s) shorted?
I found the two channels are slightly different, I think everyone else has as well...
Looks interesting.
_-_-
Hi Bear,
the sweet spot of my unit is at 500mVrms.
So I am always adjusting the buffers to run the QA in the sweet spot.
The shown results with filter and buffer are looking identical between
Output gain =1 (input gain also 1) up to output gain =7 (input gain adjusted to 1/7). When the output buffer has to deliver signals above 5Vp, then the distortion starts to rise.
Attached the loop back results of the QA400 alone - done some weeks ago, right after I received the unit
Also my QA has small difference between left and right.
Overvall the right chanel is slightly better, that's why I am using it.
Will look into 192kHz next....
the sweet spot of my unit is at 500mVrms.
So I am always adjusting the buffers to run the QA in the sweet spot.
The shown results with filter and buffer are looking identical between
Output gain =1 (input gain also 1) up to output gain =7 (input gain adjusted to 1/7). When the output buffer has to deliver signals above 5Vp, then the distortion starts to rise.
Attached the loop back results of the QA400 alone - done some weeks ago, right after I received the unit
Also my QA has small difference between left and right.
Overvall the right chanel is slightly better, that's why I am using it.
Will look into 192kHz next....
Attachments
We can use a stripped down version of the Shibasoku autoranging circuit. Its self contained and reports back to the host what the gain setting is. It has much finer stepps than would work with the QA400 (1 dB vs. 20 dB) so the stripped down version should be simple. It could report back to the host what the current gain is to get corrected display settings but that adds considerable complexity.
My suggestion would be the following as giving the most flexibility-
300 mV, 3V, 30V and 300V full scale. This would make the display easily understood. The interface box would get more complex and expensive with 6 relays and protection circuits etc. The input connectors will be an issue etc. (All why I used the interface cable to deal with the issue).
On the low pass filter at the input. I realized the industry standard is 20 KHz low pass, not 200 KHz. Also the ADC has enough dynamic range that it probably would not be an issue but I'll try it with a switching amp soon to confirm. However if you want to measure a non-oversampling DAC with no reconstruction filter there could be lots of issues.
Just want to measure amps/preamps which have higher output voltage potentials than the QA400 can handle directly. So an autoranging circuit in front of it will be auto-protecting.
Maybe 100mV, 1v and 10 and 100v RMS would make scaling and display easy to interprete?
If the SkibaSoku autoranging is simple and easy, then I'll go for it --- can you provide a schematic?
THx-RNMarsh
Maybe 100mV, 1v and 10 and 100v RMS would make scaling and display easy to interprete?
If the SkibaSoku autoranging is simple and easy, then I'll go for it --- can you provide a schematic?
THx-RNMarsh
Hi Bear,
the sweet spot of my unit is at 500mVrms.
Attached the loop back results of the QA400 alone - done some weeks ago, right after I received the unit
....
When I did a parallel test and compared the harmonics measured by a much, much more expensive and accurate units, the harmonics shown by the QA400 could not be used below -100dB. They were very different. I used the AP-2722 and the ShibaSoku 725D for reference. Only when you drop the input level via a notch filter circuit did the data shown below -100dB begin to measure what the ref units data measured.
THx-RNMarsh
I have a half-formed project using a Boonton autoranging input card in a stand-alone fashion with a small micro and a couple of detectors and comparators. Which actually are on another Boonton card.
Jan
yes, something like that also sounds like it would be an asset to any ADC used for test/measure.
Is it simple and easy, circuit-wise? Any programming needed would be a killer.
THx-RNMarsh
I don't think so. The QA400 gives a ratio of THD versus the input level. If the autoranger makes sure the input level is around 500mV, that will work OK.
The other thing is that you want to know at which level this is the THD, so the autoranger would have a small LCD display that shows either the actual input level or the attenuation in dB, selectable.
Jan
The other thing is that you want to know at which level this is the THD, so the autoranger would have a small LCD display that shows either the actual input level or the attenuation in dB, selectable.
Jan
Excellent idea and points, Jan.
One question/consideration. How best do we make sure that the auto-ranger does not briefly allow a larger-than-safe input to the QA400? Do we follow it by a safety clipper that hopefully introduces no distortion when well away from clipping? Or do we somehow always "start it" on the highest range (largest attenuation) and allow the auto ranger to work its way down? If so, how do we know when to reset it to its highest-range starting point? And what if after it has ranged there is a sudden increase in input level that would exceed the safety value? There may be some interesting challenges here unless I'm missing something.
The safety clipper is more bullet-proof, but might be a challenge to make non-intrusive distortion-wise. If we use a pair of reverse-biased back-to-back diodes that are unity-gain bootstrapped with the signal, maybe this works. I'm thinking of something like the flying Baker clamps described in my book.
Another possibility might be a shunt JFET attenuator that is controlled by a fast circuit that monitors the level into the QA400.
How long a large over-voltage transient (if any) can the QA400 take before damage has been done?
Cheers,
Bob
Once you have the autoranging you also need protection or the input will fry before the autorange ranges. That circuitry can limit the noise and distortion of the input circuit.
My contra-argument for the autoranging is that the high voltage sources will be banana connections and the low level sources will be on RCA, XLR or minijacks. This was the origin of the adapter cables with attenuators in them.
My contra-argument for the autoranging is that the high voltage sources will be banana connections and the low level sources will be on RCA, XLR or minijacks. This was the origin of the adapter cables with attenuators in them.
The best solutions I have seen use diodes and big resistors or light bulbs to handle to overload (and fuses if the autorange doesn't range fast enough. . . ) There are examples of this in the input circuit of the Boonton and the Shibasoku. I have others but they all use significant real estate. The good news is that they are essentially invisible in terms of distortion.
If you look at the size of the Boonton card you will get a sense for what it takes for a differential autoranging circuit. There is room for the logic on the cards but its still a large piece of hardware.
..great..
So it is not better than my old set up with the creative sound card EMU tracker pre.
During buffer design I noticed another obstacle: With inverted signals (measuring only the output buffer), the QA400 starts to measure a 2nd harmonic with approx. -110db, which is gone when going through another inversion (input buffer)...
@Bear:
...sorry some more patience please. I am right now suffering from empty batteries... Sounds like a lame excuse - I know.
I'd set up a window comparator for say 500mV +/-25%, plus a muting relay between the autoranger output and the QA400. You'd only unmute if the signal is within the window. I think you can set up the software such as to make it foolproof, like 'always muted unless...'.
Don't forget that the QA400 input seems pretty robust, I've seen lots of red LEDs indicating overload with no ill effects, but the 500mV is optimum for best performance, not that anything above that is destructive. So even if you want to have a safety clipper, it can be way above the optimum input level, like 3V, and pretty harmless wrt 500mV.
jan