New user here - just waiting to receive my unit. Meanwhile, I am preparing my lab. I need to measure AC voltages at about 100Vrms and thus need an external differential attenuator to protect the inputs of the device. To be able to build one I need the following info:
What are the single-ended and differential input impedances at the inputs of QA403 for the low-scale input settings (0, 6, 12, and 18dBV)?
What are the single-ended and differential input impedances at the inputs of QA403 for the high-scale input settings 24, 30, 36 and 42dBV) with the internal attenuator activated?
I was not able to find explicit information in the user manual - or any other documentation about this. I need the exact info and not any speculative data. Why is this not clearly stated in the manual? This is some basic stuff that must be mentioned in the specifications! Please add it there. You already explained in the text some parts of the input structure. A simple input schematic or block diagram would be so much more informative. The input impedance is a recurring question and must be known. Otherwise, it is difficult to build an external signal attenuator, and at the same time preserve the precision of the unit.
BNC Inputs
The 4 audio signal inputs pass through a 4.7uF series capacitor, followed a series 470 ohm resistor, and followed
by a shunt resistor divider with a total impedance of 100K ohms.
Thanks for chiming in. Yes, I found that info embedded in the text; thanks for pinpointing it. So far, so good. But is this still true for the higher input voltage range (24-42dBV) when an additional input attenuation is activated? This is not explicitly stated anywhere.
A small block diagram in the manual of the input structure would be as useful, as the block diagram for the output structure (albeit it omits the 100-ohm output resistors).
Hi @Christian, yes, it should be true for all input ranges. The input divider is a 100k impedance (around 93K and 7K). When the atten is off, you are taking the signal across the entire divider into a non-inverting buffer. When the atten is active, you are taking the signal across the 7K into the non-inverting buffer.
Understood. My external divider will then work over all ranges, selected internally by the QA403. As long as I take into consideration the loading (100k + 470ohm) that the QA403 presents at its inputs - and scale my external divider accordingly to preserve the absolute precision of the measurements.
I’m also going higher voltage measurements with the QA403, so I’d be interested in hearing about your progress, pleased do post your news.
If of interests, here is my test fixture with attenuation on the front end. It’s aimed at impedance measurements, left channel measures voltage across the DUT and the right channel measures voltage across a current sense.
I finally got my QA403 and did some input testing. The inputs seem to be 100kohm through the whole input level settings, parallel with something like 100pF.
This is what I put together, the simulation looks very good and I will soon build it. Opting for 3 different settings
0dB (bypass - for the best SNR and lower input voltages)
-6dB attenuation when the input can withstand 112Vpp
-10dB attenuation when the input can withstand 177Vpp
R1/R2 compensates for the input loading of the QA403, making it easier to dimension the other resistors. R3/R4 attenuate the -10dB setting alone, together with the R7/R8 shunt resistors. The -6dB setting connects R5/R6 in parallel with R3/R4, producing a precise -6dB attenuation. It would have been possible to have the switch arranged so that the attenuation order goes 0dB/-6dB/-10dB but this arrangement would have caused higher power dissipations in the resistors. The attenuator should be built differential as in the schematic, to preserve the best common mode rejection of the QA403 the whole way to the speaker outputs in the PA.
This should work with all power amplifiers unless we are talking high voltage transformer-driven PA systems.
I will build this attenuation directly in my 300W/600W dummy speaker load (4/8/16 ohm). I will post some pictures and test results later. You can use 1% resistors and hand-pick the most precise ones, to preserve the best accuracy. Do not use resistors with a higher temperature coefficient than 50ppm! Good thick film types are recommended.
