Literally looking at my precision 10 ohm power resistor as I read your comment. Something is funky somewhere.
Mark
Same situation. Unplugged the USB hub and plugged it back in, same. I will chase the computer connection shortly. Helping the oldest son make a set of Kitchen Cabinets. The perils of growing up in a Cabinet shop!
Mark
Good luck with both !
1 Like
I am having a similar issue with noise and incorrect measurements. The measurent here is of a fixed resistor. Measured with a 6.5 digit multimeter with Kelvin leads I get 8.095 Ohm. With the resistor directly hooked up to the “Output” of a QA461 driven by a QA403 I get this:
Voltage at the QA403 is given as 4.8V. Powering it from the laptop or through a powered hub doesn’t change anything. Neither do attenuator settings or changes in drive level. Changing the FFT size has a slight effect on the wiggles below 30Hz, rest remains unchanged. Different BNC cables, taking off and putting back on the terminating plugs on the negative inputs of the QA403 has no effect.
1 Like
Can you post the left and right channel voltages from the main plot window? It looks like one or both of the signals is getting small and adding noise when the complex division is done.
Also, what is the output level using, could it be too high and distorting the op amp?
It would be good to post the settings you use for the impedance measurement.
1 Like
Hi Dan, thanks for getting back. Here is a screen shot that shows my settings. Everything below 1k is a straight line.
-20dB doesn’t look like an overloaded input?
The resulting impedance plot is that:
Changing the DC Resistance value in the the settings doesn’t change anything, neither does a change of the attenuator setting or anything else.
1 Like
Hi THDaniel,
I found Matt’s post explaining what is going on, it’s just above in this thread actually, Speaker Impedance with QA403 + QA461 - #94 by matt it looks like the same thing to me.
I think this is the key part:
I think the thing to try is driving the load harder to make the current sense measurement less noisy.
You can also try sampling at a slower rate and picking a windowing function that might help.
1 Like
Output level -0dB:
Output level -40dB:
I tried to switch from Flat Top to Hanning, but the boxes didn’t react.
1 Like
Judging from the Bode plot the 403 and 461 seem to be interacting well. That’s the response into an 8R1 and 4R load.
Maybe the current sense output just isn’t meant to measure speakers.
1 Like
I found some documentation/app note from Matt here https://quantasylum.com/blogs/news/speaker-impedance-measurements
He measures a 4 Ohm speaker impedance and the data looks great. There are some suggestions for using FFT Size that might improve things: using a chirp that lasts 1 second.
Key Point: Use a 64K or higher FFT at 48Ksps. This will yield a chirp that is just over 1 second, and it’s accuracy at resonance will be within a few % of the value seen with a 128K or 256K FFT.
However there is a note about using the QA460.
Key Point: The QA460 is designed precisely to help you make speaker measurements, but you might want to increase the shunt resistor to 0.1 ohms or even 1 ohm, depending on the accuracy you need to see in the resonant peak amplitude. It’s not uncommon to see resonant peaks of 20 or 40 ohms when measuring speakers. The good news is the sense resistor doesn’t have an impact on resonant frequency–just the accuracy at peak resonance.
If you don’t fancy swapping out the 0.01 Ohm current sense for a 0.1 Ohm you could use the impedance test fixture PCB I made. This way you can use the QA460 to drive the load but use whatever current sense resistor you want. There is a thread for that and Gerbers + BOM on GitHub Comments welcomed on test fixture PCB for impedance measurements - #11 by Dan
Finally, I’ve been making a reactive load for my guitar amplifier, and get similar data to you when I take measurements. As the impedance rises the voltage drop across the current sense gets smaller and noise appears. It seems like I too could improve these measurements!
Solid lines are my simulations, others are QA403 + QA460.
1 Like
Thanks for collecting all that. I am still a bit baffled about the 461 current sense. And I wasn’t able to replicate the measurements in the blog post. There is just too much fuzz and the values given are off by at least 5%, no matter what settings I try. I’ll hook up the impedance jig I use with REW or Fuzz Measure (I still have an older version that used to have an impedance measurement function) and see how that works with the 461. That always gives me a straight line without any fuzz at 8.0 when measuring an 8R0 resistor. If that works out nicely I’ll swap the current sense resistor for a 2R0 to replicate that 100R sense resistor that works with REW.
I took out to sense resistor and replaced it with 0,47R and 0,68R. This is the effect:
This looks a bit better, but still far from “good”. Next step is probably to take out the INA199A1.
The load to be measured was a 4R0 resistor, so this isn’t confused with the graphs posted earlier.
1 Like
@matt To understand the QA461: was it primarily designed to get Thiele Small parameters for woofers? For that it makes perfect sense.
It works, but some of the calculations are driving me a little crazy.
I have yet to use the online app by Claus Futtrup and John Candy 
This will be a baseline comparison. I can do impedance on the Speakertester Pro and on the 403/461. Because there is something that is still not providing me the same calculated parameters when I compare the two measurement systems.
Impedance should be an independent parameter, and the calculations through a rigorous derivation via Speaker Bench.
I have used the Speakertester Pro for over 12 years. Always rock solid.
The teething pains for the 403 will be over soon enough I hope.
Mark
Hi Mark. I’m not on your level, when it comes to measuring drivers. When I start building a speaker I get an impedance plot to be able to judge what the lowest crossover point would be to not kill the driver. Several steps later I check Qts of the woofer in situ.
The issues I have are not 403 related teething pains but unfortunately related to the 461. Which, as is doesn’t give me anything useful for anything else but (sub-)woofers. For that the current amplifier works great and is a perfect solution.
Yes, @THDaniel, yes, you are right. The QA461 is designed to measure larger currents. The upward slope you see in your plot is probably related to the INA199 bandwidth (I think that corner is at 70 kHz). If you measured with a 10 ohm differentially in series with the output, how would that look?
I was using my QA461 but got sick of putting multi core wire cable into this connector. It makes unreliable connections.
Does anyone do this properly with ferrules? Or how it is intended to be used? I need to prepare my cables properly!
Used correctly, Phoenix connectors are very reliable. I see them misused the most when people don’t open them fully before inserting the cable. Run the screw counterclockwise until the jaw is fully open. If you don’t, ou can insert the wire below the jaw and it won’t get tight. Ferrules can be used, but I find them unnecessary if a stranded conductor is twisted tightly before inserting. Don’t tin the wire with solder, as the solder will flow over time and loosen the connection.
Dale Shirk
1 Like
Thanks for the helpful advice. 
Hallo I am an new User of the 403 Analyzer and ordered as well the 461 Amp to measure Speaker TS Parameter with it. After Reading here i am a bit confused.
So if i use both Devices for Speaker Measurments i need to remove the 0.02 Ohm Senseresistor on that 461 Device and replace it with an 2 Ohm Resistor to get 100Ohm like with REW for Impedance.
To get the best Results ???
After the ASIO Drivers are Out may be Using REW to measure the Impedance isn,t the worst idea to get TS Parameter like i do it with my Soundcard Setup before.
SO i can Use the ARTA Box (ARTA Box 4.1 Kit - Ingenieurbüro Grafe) with both Devices as well ? Has anyone Tried it yet?
My Devices orered this day so i cant test it now but i will.
Anyway i will build me an new Front Panal to build the Devices into an 19" 3HE Enclousure may be i can add an good Goldplatet Switch and Add Some different Sense Resistors to switch between them not sure how the 461 Device looks from inside bud adding some short wire to the Switch must be possible??? I have an ELNA Input Selectorswitch here may be its an Option to use. (0.02Ohm / 1/4 Watt, 0.2 Ohm / 1Watt, 2 Ohm / 5Watt, and 20Ohm / 10Watt or so) Any advice ?