I am looking into doing some testing of a preamp that has balanced input & outputs and was wondering what solutions others may have found for this, other than making your own cables. The side connecting to the QA40x +/- inputs and outputs would need to be BNC to connected to an XLR plug or jack (A total of 4 cables would be needed for stereo testing). I saw a post about possibly using these cables on ebay:
and then I suppose you could use these to convert to XLR:
and
Thoughts ? Thanks!
Another option is to make an adapter PCB with the sockets you need, then rely on standard cables. For some reason I tend not to trust these one off adaptors and cables - probably the cables are fine it’s the adaptors.
Thanks for the idea. I think it would be easier to wire my own than build an adapter pcb. I think I have some silver microcrophone cable that was given to me that might be useable…
What I have done is get bnc cables on amazon or ebay cut in two and terminate into XLRs. Thats easy. BNCs usually require specific cables and crimp tools. And practice with failed connections that need to be tossed.
Most adapters will work OK but you need to check the connection order. There are two standards for XLRs.
My opinion is a patch panel may be the best option. I’ve made up plenty of XLR to RCA/BNC and XLR to XLR leads for previous analyser setups, but with balanced in and out, you need quite a few and let’s be honest, making up perfect length test leads gets old fast. And you have a rat’s nest of hanging cables everywhere which always end up tangling when you try to retrieve just the pair you want…
I’m going to make up a custom patch adaptor panel with chassis mount XLRs (male and female for in and out), RCAs and BNCs, and 4mm, with rear mounted BNCs for the 8 links to the QA-403. Probably also put shorting toggle switches for the various inputs when running SE. May even incorporate the connections/switching for the loads, depending on the size/layout. It’s a job I’ve been putting off for too long.
It appears that there is a similar thread, but more focused on differential cables. I do like @1audio 's suggestion about cutting one end off of bnc cables and soldering them into XLR adapters. Sometimes a local electronics surplus place has a good deal on bnc cables- now will look for the xlr connectors. I have ran of room on my pomona wire/cable holder so appreciate @restorer-john 's preference for a patch panel. Most likely once I am done with them I will bag them up and put them with other cable sets that I do not use that often.
Howdy,
The cable you linked will work for differential TRS to the QA401,
If you want balanced to unbalanced there is a different cable from the same vendor I will link at the end. This vendor will do custom as well cheap, so if you want the same thing but with XLR just ask and I am sure they will do it. The ones I got we custom. Easy process, still very cheap price.
I bought 4 of them and the performance was stellar! It dropped the 60Hz by over 12dB from using regular single ended BNC cables and the rest of the noise floor went way down. First picture is BNC loop back with terminators, 2nd picture is using the cable linked.
My goal is to test balanced to balanced. I am looking a building my own at this point, but have to see if I have enough cable to do it…
I wanted to update as to what I ended up doing as well ask a question about my results. I had just the right amount of 2 conductor braided silver “microphone” cable to cut into 8, 3’ sections. I only used one conductor from each cable and the shield to each bnc adapter, which were these:
and for the xlr connectors I used these:
Was not able to get both the cables through the XLR’s connectors opening at the end, but did find a workaround. Takes about an hour to make one channel’s input/output cable (there are 4 of them total, of course).
Once I had one cable set made, I connected it the left channel and did a loop through measurement and got the trace on the left. Then I did a “normal” unbalanced loop through on the same channel and got the trace on the right. The atten was set for 12dB, sample rate=192k, 128k FFT and HANN windowing:
A little improvement in the SNR and THD+noise, but nothing drastic- at least for the kind of things I look at. Being a “newbie” into the world of balanced measurements, I get that the voltage is doubled and the gain is up by 6dB, but does that mean I subtract 6dB from the gain readings I get when measuring a preamp using the balanced inputs and outputs? Do I need to tell the QA40x that I am making a balanced measurement? I saw somewhere about an impedance test that allows for a balanced measurement, but that was it. It’s possible there is a setting for it that I have never noticed…
As far as I can see, there is no using balanced output button to check for the generator. It’s not the analyzer that’s at fault- it’s measuring the correct voltage on the differential inputs.
Put your ACVM/DMM over the ‘hots’ on the gen output (in idle) and you’ll see double the voltage indicated on the gen1 amplitude panel when utilizing differential outputs.
So, to answer your question, if your preamp say has a 2V balanced sensitivity for rated specs, set the gen to 0dBV (1V) and your gain numbers will be correct.
Basically, there just needs to be (somewhere) software check boxes for balanced/SE on both inputs and outputs. Maybe it’s hidden someplace already, but I can’t find it.
@restorer-john Thanks for the reply. I checked the preamp specs that I will be measuring (McIntosh C48 
) and it provides separate ones for balanced and unbalanced inputs. THe THD & Freq responses are the same, but max input and output levels are different of course. I have two more bnc connectors to mount and my cables will be done and I can start testing it. I am looking at purchasing this from a friend so will be testing a lot, which keeps me off the streets at night 
You can use the offset function too. Offset by 6dB and the the gain numbers will be right and the input and output numbers. Just check the actual signal amplitude going into the preamp and what is coming out with a DVM as the offset panel is a bit head scratching when you first look at it.
Thanks @restorer-john. Easy enough to do. Will try it out tomorrow sometime- will be checking out my new cables then with the preamp. Just a few quick tests to see how they work…
As a follow-up, my balanced cable set seems to be working ok. Here is a measurement I did on a preamp, 1st with the balanced inputs/outputs (left plot)and then with the unbalanced inputs/outputs (right plot). I did not input any offsets, but I did have to increase the gain on the preamp when I went from the balanced input to the unbalanced input.
The balanced input and output looks cleaner, although the unbalanced input’s SNR is better…
BTW, this is the best preamp as far as thd/snr/distortion that I have measured to date.
The C48 has the following specifications.
The input sensitivity for rated output and publsihed specs is 450mV unbal and 900mV bal for 2.5V and 5.0V output (rated output).
You’ve fed in 2000mV unbal and 4000mV bal. S/N will obviously be helped by a hot signal, but it is very close to spec and seems to have a fair bit of headroom in reserve, although it’s a relatively low gain preamp.
When testing preamps to published specs, the volume control should be set to 0dB (full volume with analogue preamps) and the input signal adjusted.
Glad your cables worked out. I’m still putting off building by patch bay (it’s way too hot for the workshop)
@restorer-john, thanks for your commentary on this! I had printed the spec sheet out so I have it at hand when I start doing measurements- which I will start in earnest once I finish testing my Pioneer qx8000 quad receiver for an upcoming video (plus I am going to try to sell it). The C48 max gain from the line output looks like 15dB, which seems pretty good compared to other preamps I have tested. I always start off with 0dB gain and 2V in, and may add another voltage testpoint if it is specified. I will also do the test with 450mv/900mv and adjust the gain to get either 2.5v/5v, and it looks like the A weighting filter will be on for those. The C48 has a DAC and I will do some limited testing on it- basically using a very “high resolution” 1khz .wav file generated with Audacity, such that the QA403/2 can read it and compare that with what comes out of the C48’s line output (both balanced and unbalanced). The problem is that I cannot go directly to the C48 DAC input as the 1khz .wav file must be played through my NAS drive through a Wiim Mini streamer and feed it’s optical output to the C48’s optical input (I think the wiim mini’s optical output did a pretty good job when Amir tested it). I can also play the .wav file with Audacity and select my computer’s optical output and see if the results are any better. I don’t think I would be able to use the QA40x to analyze a .wav freq. response wave form- not sure if I can generate one with Audacity (or REW)…thoughts? For reference, this is what the Audacity generated tone .wav file looks like to the QA402 (it is 6minutes long:
Funny, I have a QX8000a too. Picked it up over a decade ago. Too many projects, not enough time. 
Why don’t you use the USB input on the C48 and test using PC mirroring?
Your’s is the “a” version and mine is not- one difference is where you MIC is my QUADRALIZER button is. I actually bought a Quad Panasonic 8track that I had hooked up to it until I got a great deal on a Marantz quad so the Pioneer had to go. I really do not know anything about the PC Mirroring function, so thanks for suggesting that. I did connect the C48 up to my PC’s usb port, but had to download the usb drivers from McIntosh to have it seen. THen I played the tone through Audacity, selecting the C48’s USB input and looked at the main output (unbalanced) with the QA403 and got this:
Not very impressive, IMHO. I am guess part of the problem may be just my PC, but I will certainly explore PC mirroring.
@Restorer-john- I got around to measuring the C48’s USB input using the QA403 in PC Mirror mode and this is what I got:
This gives better results than I got previously. It did not work when I tried an Expochirp FR. What I am not sure what the QA403’s generated output looks like (have not used the Display OUTPUT, so maybe that will tell me?) I do have another DAC that I can test to compare it to (SMSL SU-1), which I will try as well.
Reduce the output level in Sys:Gen1 to -0.2 dbFS (shows -0.2 dBV) and see what you get. There might be a bug in the implementation of Mirror that causes clipping when the level is set at 0.
