Hi ! thanks a lot for the very valuable advice. I will read the thread before asking anything else ...
I already saw this ...
i have also a Tascam uh 7000 (stored away and not accessible now as i am on the road) and i got this
the ultrasonic spikes went away bypassing the internal smps (a known issue) and using instead an off the shelf Sony 12VDC smps
Maybe i should go and get this Tascam
But as i said the very important thing i have learned is to look at the ADC chips ... and that their noise determines the overall noise of the card. Thank you All sincerely. Issue close. I will follow the thread mentioned with great interest.
Kind regards, gino
Last edited:
Current! If the device needs more than about 500mA then you are in trouble if only using USB power.
Tony.
The ADC is NOT the final frontier. For noise measurements, one uses an
external preamplifier with 50 to 80 dB gain and the line input. That masks the ADC
noise completely. The ADC needs to sport a flat frequency response and that's it.
ADC dynamic range is needed only for distortion measurement. Dynamic range
is the weak point of my Agilent 89441A FFT analyzer with its 12 Bit 24 MHz ADCs.
It gets problematic below 100 Hz when its excessive 1/f noise finally wins against
the added preamplifier gain.
But a sound card does not have a 12 bit ADC.
The final frontier is the noise of the preamp input stage, ground loops and
bad shielding. Local batteries in the preamp help a lot.
I usually have the preamp in a Hammond box with batteries inside and the DUT
in a second Hammond box.The 2 boxes are connected by a short piece of
semi rigid coax cable and nothing else, and everything together is in another
large metal box.
When the 2 Hammond boxes inside touch in addition to the semi rigid, you see
additional signals immediately because of the 15 cm ground loop.
external preamplifier with 50 to 80 dB gain and the line input. That masks the ADC
noise completely. The ADC needs to sport a flat frequency response and that's it.
ADC dynamic range is needed only for distortion measurement. Dynamic range
is the weak point of my Agilent 89441A FFT analyzer with its 12 Bit 24 MHz ADCs.
It gets problematic below 100 Hz when its excessive 1/f noise finally wins against
the added preamplifier gain.
But a sound card does not have a 12 bit ADC.
The final frontier is the noise of the preamp input stage, ground loops and
bad shielding. Local batteries in the preamp help a lot.
I usually have the preamp in a Hammond box with batteries inside and the DUT
in a second Hammond box.The 2 boxes are connected by a short piece of
semi rigid coax cable and nothing else, and everything together is in another
large metal box.
When the 2 Hammond boxes inside touch in addition to the semi rigid, you see
additional signals immediately because of the 15 cm ground loop.
Last edited:
Very interesting indeed and thanks a lot for the kind advice. I guess for DUT you mean Device Under Test ? therefore your measurement chain is DUT > preamp > Agilent 89441A FFT analyzer ? i see that the Agilent 89441A is a very professional unit. Did you try a usb sound card + pc approach maybe ? or better, why did you prefer the analyzer approach ?
Speaking instead more in general of noise, i am extremely interested also in power supply noise measurements ... for instance out of some dc-dc converters ...
For instance i am waiting for one with usb out up to 5A that takes at the input a wide range of dc voltage. I could even use a 12V sla battery for instance.
But i do not know what level of noise the converter itself will generate.
ICQUANZX DC 6V 9V 12V 24V a DC 5V 5A Buck Modulo convertitore, 9-36V Step Down a USB 5V Transformer Dual Output Voltage Regulator Board: Amazon.it: Elettronica
Thanks a lot again, gino
Last edited:
Illustrated by Morrison in "Grounding and Shielding Techniques in Instrumentation", V 1->4
Attachments
Good low noise ADC usually goes hand-by-hand with good DAC. Together they make near top card, so it costs not less then some amount of money. I can't unfortunately recommend exact USB card (I used to use internal PCI cards). There were good Asus cards (like my Xonar STX, but USB) but they are gone from market now. Maybe, there is something at aftermarket?
You can google for "best external usb sound card", and search for ones with about 120 dB signal-noise ratio of ADC - it has to be definitely good ADC.
Last edited:
- It is not bad. You have a choice - either use this Tascam (for free) and gain experience, or spend at least $200-$300 for something "near top".Maybe i should go and get this Tascam
I recommend first way.
Last edited:
Yes, DUT means device under test.
The 89441A was really made for developing cell phones. The baseband unit has
10 MHz bandwidth and can demodulate all the fancy modulation schemes used
by the networks. But people don't get it that these features are built on a
wideband FFT analyzer. There is also a tuner unit that expands that to 2.7 GHz,
so you can receive signals from the air interface.
But it does not know the tricks for G5, so the cell phone manufacturers needed
new instruments. I think that mine was pre-owned by Motorola. Because of the
need to replace them they were quite cheap.
I once tried a Terratek USB sound card without much success under Linux.
Did not work for me, and I also needed >= 1MHz BW.
I have checked some lab supplies. Some are ugly.
< http://www.hoffmann-hochfrequenz.de...rements_On_Some_Laboratory_Power_Supplies.pdf >
cheers, Gerhard
Hi and sorry but i must have missed this advice. Which 2-300 USD unit would be near top ? but before you answer i have found a usb sound card from Behringer very promising ... said to have a very low noise
Behringer U-PHORIA UMC202HD
Hi thanks a lot for the very helpful advice. Did you check it for noise maybe ?
Hi Nicolai, very kind of you but i am not ready to do tests. I am still in the learning phase I need more information And i am evaluating options like using a mic preamp powered by batteries to increase the noise before measuring it. One like this one for instance
AD828 Stereo Dinamico Microfono Preamplificatore Consiglio MIC Preamp Modulo DC 3.8V 15V|Pezzi di ricambio e accessori| - AliExpress
Not ready. Thanks a lot anyway
Regards, gino
.
Very interesting. A much more complex and versatile unit. I like instruments even if i am not able to use them. The look i mean. Very fascinating.
Drivers can be an issue i guess ... the one i am using was about to be thrown away in the garbage bin in a little shop selling used things. I am just using if for listening some music during working time ... relaxing music that keeps me calm
i am not an expert but did they meet the specs declared by their manufacturers ?
Anyway i have shifted my attention to signal preamps. As you very well say they can be battery powered and have a very low noise floor. At that point the choise of a usb card may be less critical ?
Thanks a lot a gain
Gino
I haven't got one (no stock), but there are some measurements here MOTU M4 Loopback Measurements | Audio Science Review (ASR) Forum
Not sure if -140db meets your needs, I've seen better posted, but generally it was PCI cards, or very expensive interfaces. I was shown a most impressive USB interface here on diyaudio a while back, but I don't remember what it was, just that it was VERY good, and VERY expensive!
edit: I remembered saying "I stand corrected" so search found it Got Focusrite Scarlett 2i2 (2nd gen) for measurements. What's next? It had to be an RME
Tony.
Last edited:
Hi i have found a video mentioning the very low noise of the mic preamp
But with a dedicate signal preamp also the line in can be used i guess
Behringer UMC202HD review (with noise measurement) - YouTube
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.