LOL 🤣 Yeah... I'm about 20.4 dB younger than that. Most don't expect that either. 🙂
Perhaps I should have said, "when I bought my APx525 years ago, ..." I didn't expect to get misunderstood that way, though I can see how you got there.
Tom
Perhaps I should have said, "when I bought my APx525 years ago, ..." I didn't expect to get misunderstood that way, though I can see how you got there.
Tom
QA403 is only single ended, correct ?
We want to have balanced as well, followed PMA and use this as a balanced system :
www.audiosciencereview.com/forum/index.php?threads/topping-d10s-e1da-cosmos-adc-as-a-measuring-system.34723/
www.diyaudio.com/community/threads/autoranger-for-soundcards.299635/post-7474015
Patrick
We want to have balanced as well, followed PMA and use this as a balanced system :
www.audiosciencereview.com/forum/index.php?threads/topping-d10s-e1da-cosmos-adc-as-a-measuring-system.34723/
www.diyaudio.com/community/threads/autoranger-for-soundcards.299635/post-7474015
Patrick
Balanced using BNC connectors.
I agree Tom, wouldn't use a separate DAC and DAC. I'd want something on the same clock. Hence why loopback performance is important for me.
Things like FFT and FSAF and very sensitive to clock drift. The auto ranging function very handy, one would be forever wiggling around knobs on a standard consumer grade audio interface - you spend more time calibrating than actual testing! In fact, I wonder whether people are actually maximising their signal to noise ratios when doing the test... this a culprit of many acoustical measurements that at first glance, appear to show a very poor DUT (ask me how I know)
I agree Tom, wouldn't use a separate DAC and DAC. I'd want something on the same clock. Hence why loopback performance is important for me.
Things like FFT and FSAF and very sensitive to clock drift. The auto ranging function very handy, one would be forever wiggling around knobs on a standard consumer grade audio interface - you spend more time calibrating than actual testing! In fact, I wonder whether people are actually maximising their signal to noise ratios when doing the test... this a culprit of many acoustical measurements that at first glance, appear to show a very poor DUT (ask me how I know)
Viktor has an oscillator with a balanced output? The ones I have have only single-ended output. It's also floating at VCC/2 which is a bit inconvenient at times.
Yeah, yeah. Cheap jab. It's an annoyance I'm willing to deal with in exchange for the low price.
Tom
How does the clock drift come across? Poor side lobe attenuation?
Yeah. "Calibrating". Most don't have the equipment to do that properly. That's another advantage of proper test gear. You know that when you set it to provide 1.00 V out, you get 1.00 V out ... unless you use the balanced out on the QA403. Then you get 2.00 V out. 😉
Yeah, yeah. Cheap jab. It's an annoyance I'm willing to deal with in exchange for the low price.
Tom
I totally agree, Tom. All it needs is a checkbox box that asks whether you are using SE or balanced and sets the levels accordingly. It could be on the gen set screens or global for the test session.
From what I have seen, Keysight is no longer following "the HP way". Since they went to the far east, software is poorly written with bugs. They take forever for the instruments to boot from Windows. QC issues on top of that. Of the last 7 Keysight instruments I bought, 5 had issues, 2 are permanent design issues that limit their use, a couple corrected with firmware patches and my 6812C AC power supply died after less than 10 uses at light load.
I bought Keysight simply because I was use to the excellent design and reliability since I bought my first 34401A new, and before when I had a 4261A earlier in time. Other brands failed to support their equipment, HP did. Today my views have changed.
For the silly high amount of money they ask for test equipment these days, and the poor build quality, poor software and limited support, they need to be educated about the concept of value. The shielding is bare minimum. Accuracy specs aren't much better than much older instruments unless you hiut high frequency (then the prices go stratospheric).
I feel strongly that AP have followed the same route. Failure to support older equipment is a policy decision, not a technical issue. They shouldn't be rewarded for this. As for software .. yes. The seminar I attended was pretty much all about the software at $3,500 asking (then). Sorry, that software isn't worth $3,500.
Our modern world is a rip-off.
I bought Keysight simply because I was use to the excellent design and reliability since I bought my first 34401A new, and before when I had a 4261A earlier in time. Other brands failed to support their equipment, HP did. Today my views have changed.
For the silly high amount of money they ask for test equipment these days, and the poor build quality, poor software and limited support, they need to be educated about the concept of value. The shielding is bare minimum. Accuracy specs aren't much better than much older instruments unless you hiut high frequency (then the prices go stratospheric).
I feel strongly that AP have followed the same route. Failure to support older equipment is a policy decision, not a technical issue. They shouldn't be rewarded for this. As for software .. yes. The seminar I attended was pretty much all about the software at $3,500 asking (then). Sorry, that software isn't worth $3,500.
Our modern world is a rip-off.
Several years back I was on the search for a small-size, transportable, affordable and capable replacement for my AP SYS2322 boat anchor, for "bread-and-butter" measurements on line-level circuits mostly.
Luckily, at the same time the RME ADI-2 Pro came on the market and it was the perfect match from the beginning, in combination with REW. Later on, RME expanded the features and provided DC output and DC input (with internal coupling caps removed). By this, I can even power some low-power DUTs with +-12Vdc from the phones output and measure DC voltages. And there is the PC remote control app which is very handy (albeit only working on modern PCs, under Windows).
REW feature set has also been expanded vastly in recent years.
------------:-----------
As for the sotware/GUI discussion, we all know that @tomchr hates LTspice UI ;-) ... while I love it, for me it is the fastest and most intutive schematic editor ever. I have to mention that I' sort of icon-blind but can well remember short-cuts or function keys etc. Also I hate to use/move the mouse for anything else other than actual drawing of a schematic, layout or 2D/3D-CAD because tendonites quickly raises its ugly head with too mouch mouse action.
Luckily, at the same time the RME ADI-2 Pro came on the market and it was the perfect match from the beginning, in combination with REW. Later on, RME expanded the features and provided DC output and DC input (with internal coupling caps removed). By this, I can even power some low-power DUTs with +-12Vdc from the phones output and measure DC voltages. And there is the PC remote control app which is very handy (albeit only working on modern PCs, under Windows).
REW feature set has also been expanded vastly in recent years.
------------:-----------
As for the sotware/GUI discussion, we all know that @tomchr hates LTspice UI ;-) ... while I love it, for me it is the fastest and most intutive schematic editor ever. I have to mention that I' sort of icon-blind but can well remember short-cuts or function keys etc. Also I hate to use/move the mouse for anything else other than actual drawing of a schematic, layout or 2D/3D-CAD because tendonites quickly raises its ugly head with too mouch mouse action.
No, the supply needs to float and the two shunt regulators establish a ground between them. A single ended supply, even a stack of 9V batteries works well.
Clock drift will distort harmonics on a high res measurement when the harmonic frequency drifts out of the particular bin. You can get unrealisticly low harmonics.
UI- I was working with a group on a major UI project. Everyone had their own ideas. However the only intuitive UI is the nipple. . . everything else is learned. IBM had to deal with this nightmare very early in the computer days. They defines the title bar, pull down menus and their order. It has stuck for the most part for 50 years. And the UI of a printed book replaced scrolls 1200 years ago and we still use that form despite efforts to replace it. I find aspects of the AP interface obscure but I can say the same for almost all test instruments. (Even Exel is impenitrable past a certain low level) It took me a while to master an HP VNA, and lots longer for a Tektronix eye pattern and bit error rate test. And the Keysight version is very different. Its just the nature of this stuff.
I find the QA products pretty quick to get up to speed with on most basic tests. If you need to go super deep on harmonic distortion there is the Cosmos https://e1dashz.wixsite.com/index/cosmos-adc stuff which costs less than an AP cable set.
You remember the schematic better than I do. Either way, not the most convenient. I've used a stack of 9 V batteries. That works well as you point out. As do a couple of power tool batteries.
Tom
I had no issues getting the testing going. It was all the work related to getting the test results ready for publication that was a pain.
That may be. But there's an entire field of psychology that deals with how we can make machines work well with humans. It usually falls under Human Factors Engineering. We can study some of the mistakes others have made and design something that works well.
Have a look at REW. That's pretty intuitive to use in my (not so) humble opinion.
LOL <snort>. Yeah. They do indeed offer amazing performance at their sweet spot. For less than I spend on postage every month. 🙂
Tom
REW is quality software addressing needs of both occasional users ("find icon and click" mouse style of operation) as well as everyday power users (sufficient set of keyboard shortcuts removing a lot of extra effort).
Coming back to this, REW basically covers it. If you wan to go fancy with real/imag or polar view, Smith chart etc, you can export the Bode plot data, patch it a little to .s2p format (Touchstone format) and use any of the Touchstone viewers for the RF-style displays of S21 (transfer gain).
Watching your video Tom.
Interesting to See Software options out there and being used in real time.
Giving the human to machine interaction.
I havent really done software based audio since 2000/2003
Was using MAudio 10/10 soundcard so 10 in 10 out 24bit 96 kHz
Be interesting what the measurements on that old guy were.
I was constructing my own software with Native Instruments Reactor.
So basic test Oscillators, FFT analysis and did my own active crossovers or " DSP "up to 5 way
Since the filters and delay line options are rather extensive back then.
Just the human to machine time was rather long. I had to virtually wire all the macros.
But could design my own control panels.
Anyways blah blah.
My interest is I wonder how the Behringer U phoria UMC202HD holds up.
It has beaten a few higher priced units with other more music related measurements.
24 bit / 192 kHz 2 in 2 out Balanced In
Noise floor is -129 dB and the channels believe around 20 dB range
So should be well over 7 volts without clipping.
Be awesome to see what distortion is with all that nice equipment you have. Retail around 69 dollars
Interesting to See Software options out there and being used in real time.
Giving the human to machine interaction.
I havent really done software based audio since 2000/2003
Was using MAudio 10/10 soundcard so 10 in 10 out 24bit 96 kHz
Be interesting what the measurements on that old guy were.
I was constructing my own software with Native Instruments Reactor.
So basic test Oscillators, FFT analysis and did my own active crossovers or " DSP "up to 5 way
Since the filters and delay line options are rather extensive back then.
Just the human to machine time was rather long. I had to virtually wire all the macros.
But could design my own control panels.
Anyways blah blah.
My interest is I wonder how the Behringer U phoria UMC202HD holds up.
It has beaten a few higher priced units with other more music related measurements.
24 bit / 192 kHz 2 in 2 out Balanced In
Noise floor is -129 dB and the channels believe around 20 dB range
So should be well over 7 volts without clipping.
Be awesome to see what distortion is with all that nice equipment you have. Retail around 69 dollars
Ah. Thank you. I've seen that with Victor's oscillator too. At least the two I have have quite a bit of frequency wander/ELF jitter.
Tom
I'm not sure any software is. Yet, Cadence charges something like $5k/seat per year for Altium.
Ship one to me and I'll measure it for you. If it's any good I'll make a video about it. I'll ship it back after.
Tom