logo elektroda
logo elektroda
X
logo elektroda
Advanced Search
logo elektroda

A way to get the Behringer UMC202HD card

tytka 14733 47

TL;DR

  • Modifies the Behringer UMC202HD rev.1.12 audio interface into a direct line-input version for audio measurements by bypassing its factory input amplifiers.
  • The key change feeds the inputs straight into the signal symmetrization stage on IC8 AD8694 and then the IC7 CS4272 converter, skipping the MIDAS preamps entirely.
  • The module uses two OMRON G6S-2F 5V relays, 47uF 16V capacitors, a 1206 100nF capacitor, and a 1N4148 diode.
  • The modified mode improved the test results enough to make the card suitable for audio measurements, while a switch preserves the original factory behavior.
  • It works only with unbalanced jack inputs, voids the warranty, and excessive input level in modified mode can damage IC8 AD8694.
Generated by the language model.
ADVERTISEMENT
Treść została przetłumaczona polish » english Zobacz oryginalną wersję tematu
Report a violation of the law
Reply Cool? Ranking DIY | New topic
Notify about new articles
📢 Listen (AI):
  • #31 21481308
    tytka
    Level 23  
    Posts: 737
    Help: 8
    Rate: 1828
    Board Language: polish
    Well, that's the way it is now.

    Circuit board with various electronic components. .
    Two yellow printed circuit boards with visible traces and markings on a wooden background.
  • ADVERTISEMENT
  • #32 21481576
    marweg1967
    Level 14  
    Posts: 131
    Help: 1
    Rate: 39
    Board Language: polish
    Well, and beautifully. Even the soldermask colour appropriately matched to the factory plate :) .
  • #33 21482596
    marweg1967
    Level 14  
    Posts: 131
    Help: 1
    Rate: 39
    Board Language: polish
    Aha, it may be useful for someone to know that the UMC202HD card was devoted quite a lot of space in issue 4/2025 of the monthly magazine Understanding Electronics. Modifications to this card are also described.
  • #34 21482816
    tytka
    Level 23  
    Posts: 737
    Help: 8
    Rate: 1828
    Board Language: polish
    Has anyone come up with anything more interesting than me ????
  • #35 21483738
    KWF
    Level 8  
    Posts: 16
    Board Language: polish
    It came out super, and in almost tiles I wrote on PW.
  • #36 21483797
    marweg1967
    Level 14  
    Posts: 131
    Help: 1
    Rate: 39
    Board Language: polish
    >>21482816 .

    Not necessarily more interesting, and certainly uglier ;) Essentially the same way (bypassing the noises), except that the author has also described a way to minimise the noises without modifications and to adapt this card for use as a measurement card. There is also a link to this thread and photos of the solution presented here.

    For the full availability of the magazine you will unfortunately have to wait a while yet, for now only a pre-release for patrons is available. The issues are generally released at the end of the month. Anyway, I think it is worth reading this article if anyone is interested in learning more about the card in question,
  • ADVERTISEMENT
  • #37 21483867
    KWF
    Level 8  
    Posts: 16
    Board Language: polish
    marweg1967 wrote:
    >>21482816 .
    Not necessarily more interesting, and certainly uglier ;) Essentially the same way (bypassing the skimmers) ...


    There's quite a shuffle on the web in one of the forums about the noises of this interface and how to deal with them. I dropped out halfway through reading it because I wasn't able to keep track of the changes, The changes were quite intrusive, with various variations.

    Link: https://www.diyaudio.com/community/threads/behringer-umc-202hd-for-measurements.341309/

    marweg1967 wrote:
    >>21482816 .
    Full availability of the magazine will unfortunately have to wait a while yet, for now only a pre-release for patrons is available.


    And I already thought I had missed something, because I found it possible to purchase issues 1/2025 and 2/2025. Issue 3 is only mentioned on the periodical's website and no word about issue 4.
  • #38 21483929
    marweg1967
    Level 14  
    Posts: 131
    Help: 1
    Rate: 39
    Board Language: polish
    KWF wrote:
    Number 3 is only mentioned on the periodical's website and no word about number 4.


    There is a teaser at https://piotr-gorecki.pl/ Even a downloadable "advertorial". On the cover there is a big caption "USB audio cards as measuring instruments".
  • #39 21484092
    tytka
    Level 23  
    Posts: 737
    Help: 8
    Rate: 1828
    Board Language: polish
    From what can be seen in the publicly available, demonstration (abbreviated) copy of the ZE 4/25, it is not out of the question that Peter, in his deliberations, was inspired by my material posted here. He only simplified (and at the same time somewhat complicated) the shortcut of the audio track. Because instead of a complete solution as I presented here, he shortened the input path of the card by adding an additional input on the cables and an input switcher. (Hardly elegant).
    And my attempts with added (as in the article) diodes (1N4148), unfortunately resulted in increased distortion. So, in the end, I opted for a solution with Zenar diodes at 2.4V, because this, in my case, did not negatively affect the distortion level.

    Anyway, in my card, I also made a few other minor changes that I haven't mentioned here.
    Now the noise in my card comes out in measurements at -103.5 dBa, dynamic range 103.7 dBa, THD 0.003.

    Apparently in the full issue, there is a link to this thread. However, Mr Peter did not consult me, about the possibility of using my material, so I hope you will not see it in the full version of the issue.
  • ADVERTISEMENT
  • #40 21485005
    KWF
    Level 8  
    Posts: 16
    Board Language: polish
    tytka wrote:
    Also, in my card, I've also made a few other minor changes not mentioned here.
    Now the noise in my card comes out in measurements at -103.5 dBa, dynamic range 103.7 dBa, THD 0.003.
    .
    Could you share what else you have changed in this interface? These results look very impressive.

    tytka wrote:
    However, Mr Peter did not consult me, about the possibility of using my material, so I hope you won't see it in the full issue
    .
    Unfortunately I have also encountered such practices and they are rather contrary to Copyright Law.
  • #41 21486930
    tytka
    Level 23  
    Posts: 737
    Help: 8
    Rate: 1828
    Board Language: polish
    If I can find my notes of these additional changes to the card, I can present them here.
    I don't really remember the details of the changes and where I put the notes. It's been a long time in which I've dealt with a lot of other things.

    However, despite the corrections, it is still the UMC202HD, not the best card I currently have.
    Screenshot of test results for various sound cards. .


    KWF wrote:
    Halfway through reading I dropped out,
    .
    Halfway through? Then I guess you got to the point where my modification was mentioned there,:)
  • #42 21486993
    2konrafal1993
    Level 38  
    Posts: 2916
    Help: 581
    Rate: 744
    Board Language: polish
    I rammed the first DAC in this card and on it the results were not particularly impressive. I bought a second DAC and tested it, it reached -104dB of noise and geez I was so pleased I rammed it too. I bought another one that barely approached -101dB and it stayed that way. This reinforced my conviction that you also have to hit a better piece then you will also gain a bit. I should add that my card has only the modification mentioned in the subject. Along the way I flew some parts, it cost some time and money, the hardest thing was to find good opamps, because on some supposedly great parameters card was humming at the output, originals were impossible to get, so I had to combine and managed to catch good operational amplifiers for the output.
  • #43 21487122
    tytka
    Level 23  
    Posts: 737
    Help: 8
    Rate: 1828
    Board Language: polish
    2konrafal1993 wrote:
    This reinforced my belief that you also have to hit better art then you will also gain a little.
    .
    All in all, there's quite a bit right in those words.
    My Xonar, has supposedly worse WOs because from UTC, and my friend's Xonar has WOs from TI. And mine on the Taiwanese ones, performs noticeably better.
    Honestly, I was stunned by this, because some time earlier, I was struggling at work with the improvement of a certain device of ours, in which (due to the costs) it was decided to install replacements from UTC instead of the amplifier from TI. And with these replacements, after some time of operation, problems started to occur, the amplifiers started to excite. So previous experience indicated that the Taiwanese chips were weak. And here suddenly such a surprise.

    Going back to the UMC202HD, it once occurred to me to throw everything but the Xmos out of it and build it anew with a different ADC and DAC. But after thinking it over, I let the subject go, as it would have been unprofitable overall. Especially since I was already putting together a small measurement computer with Xonar at the time.
  • #44 21487409
    KWF
    Level 8  
    Posts: 16
    Board Language: polish
    tytka wrote:
    however, despite the corrections, it is still the UMC202HD, not the best card I currently have.
    .

    Yes, I know there are better interfaces/cards out there. I was looking at the Scarlett 2i2 before buying the UMC202HD, but the price put me off as it was out of budget for my applications. I also have a Sound Blaster X-Fi HD USB card (SB1240), but it's picky about working with newer versions of macOS (max 96kHz and 16-bit) and REW.

    Before I bought the UMC202HD I scrolled through n pages on the web, linking to this thread, so my choice was an informed one.

    Anyway, your modification on the relays appealed to me the most, for its simplicity as well as its finesse.
  • #45 21488004
    marweg1967
    Level 14  
    Posts: 131
    Help: 1
    Rate: 39
    Board Language: polish
    >>21487409 .
    I too started my search for a measurement card with Focusrite. I had already even got over the price and was really close to buying one. But I came across a video of an unsuccessful attempt to repair a Focusrite (I think it was Rafal Bielawski), which showed that it was easy to damage and difficult to repair. So I took a chance on the Behringer, well now I'm going to rework it ;) Because it performs poorly without it.
  • ADVERTISEMENT
  • #46 21488183
    tytka
    Level 23  
    Posts: 737
    Help: 8
    Rate: 1828
    Board Language: polish
    So did I. When I bought my first card, I dreamt of a Scarlett 2i2, but the budget wasn't there at the time, so I bought a Behringer. And ... First there was disappointment, and then the "fun" started, to get what I could out of it :) .
    I bought the Scarlett-ka in the end anyway, because I needed to have two cards, and a good second-hand one came along.
    And now that I have my mini metering computer with Xonar, I think I'll get rid of the Scarlett-ka. The Behringer is supposedly inferior to it, but through all that struggle with it, I have a certain fondness for it :) .
  • #47 21718910
    koleckiteam
    Level 1  
    Posts: 1
    Board Language: polish
    >>21488183 .
    I have written PW.
  • #48 21776917
    tytka
    Level 23  
    Posts: 737
    Help: 8
    Rate: 1828
    Board Language: polish
    If anyone would still be interested in a board for such a conversion, please contact me.
Reply Cool? Ranking DIY | New topic
Notify about new articles
📢 Listen (AI):
Report a violation of the law

Topic summary

✨ The discussion centers on the Behringer UMC202HD audio interface, specifically rev.1.12, highlighting its limitations for high-precision audio measurements despite its advertised 192kHz/24-bit capability. Initial tests with RightMark Audio Analyzer revealed unsatisfactory noise and distortion levels, prompting a hardware modification approach. The main modification involves bypassing the original input preamplifiers using a custom PCB with goldpin connectors, effectively shortening the audio path and reducing noise. Key components identified include quad AD8694 operational amplifiers, NJM2122, 4580 op-amps, CS4272 ADC/DAC chip, and XMOS SK1814 DSP/USB controller. Attempts to add simple limiters with diodes were made but caused distortion; a refined limiter using Zener diodes at 2.4V was preferred. Discussions also covered the variability in component quality affecting noise performance, with some DAC replacements improving noise floor to around -104 dB. Suggestions included replacing input op-amps or bypassing mic preamps entirely, though supply voltage constraints (e.g., for SSM2019 op-amps) were noted. The modification was compared to other interfaces like Focusrite Scarlett 2i2 and Creative Sound Blaster X-Fi HD, with the Behringer being a budget option that can be improved through hardware tweaks. The community showed interest in obtaining the custom PCB, with the author offering to produce and distribute boards at cost. The topic was also referenced in the Understanding Electronics magazine (issue 4/2025), which covers similar modifications. Additional advice was given on repairing damage caused by overvoltage on inputs, recommending stepwise replacement starting from logic gates and op-amps before the converter. Overall, the modification provides a practical, cost-effective way to enhance the UMC202HD for audio measurement applications by reducing noise and distortion through hardware bypass and component optimization.
Generated by the language model.
Today's popular

FAQ

TL;DR: By bypassing Behringer UMC202HD Rev 1.12 microphone pre-amps, users cut input noise to −103.5 dBA and THD to 0.003 % [Elektroda, tytka, post #21486930] “The best solution here” [Elektroda, tytka, post #18445119] Switch four capacitors for gold-pins, add two relays and a 5 V switchable board.

Why it matters: The tweak turns a budget interface into a measurement-grade tool for under €10.

Quick Facts

• 24-bit / 192 kHz USB-2.0 interface (factory spec) [Behringer UMC202HD Datasheet] • Stock noise floor ≈ –90 dBA, THD 0.02 % [Elektroda, tytka, post #18442172] • Post-mod noise −103.5 dBA, dynamic range 103.7 dBA [Elektroda, tytka, post #21486930] • Parts list: 2 × G6S-2F relays, 4 × 47 µF SMD caps, misc ≈ €8–10 [Elektroda, tytka, post #18442172] • Opening the unit voids warranty [Elektroda, tytka, post #18442172]

What causes the high noise and distortion in the UMC202HD?

The factory MIDAS input pre-amps generate strong second-harmonic content and hiss; small gain changes swing results widely [Elektroda, tytka, post #18442172]

What components are needed for the relay-bypass mod?

Parts: 2 × OMRON G6S-2F 5 V relays, 4 × 47 µF 16 V bipolar SMD capacitors, 1 × 100 nF 1206 capacitor, 1 × 1N4148 diode, SPST switch, gold-pins, thin twin-core wire [Elektroda, tytka, post #18442172]

How do I install the bypass in three steps?

  1. Desolder capacitors C81, C85, C34, C35; insert long gold-pins. 2. Solder the pre-assembled relay board onto the pins and feed +5 V from C7 pads. 3. Mount an SPST switch on the rear panel; ON = direct line, OFF = factory path [Elektroda, tytka, post #18442172]

What performance gain can I expect?

Measured noise dropped from about –90 dBA to –103.5 dBA; THD fell to 0.003 %, a 13 dB improvement in dynamic range [Elektroda, tytka, post #21486930]

How can I protect IC8 (AD8694) from over-voltage after bypassing?

Add series resistors and two Schottky or 2.4 V Zener diodes to clamp inputs; this limits level without adding distortion [Elektroda, DVDM14, #18443925; tytka, #21484092].

What happens if I still overdrive the inputs?

Excess level can destroy IC8 or even the CS4272 converter; repair then requires isolating and replacing op-amps, logic gates and finally the ADC/DAC [Elektroda, 2konrafal1993, post #21396434]

Will swapping op-amps fix the noise instead of the relay hack?

Replacing pre-amp ICs alone yields marginal benefit; layout and high default gain remain limiting factors [Elektroda, tytka, post #19217334]

Can other Behringer interfaces be modified the same way?

Users report similar noise on UMC404HD and UMC22; the relay-bypass concept applies, but component references differ [Elektroda, Piottr242, #18442742; indianka, #18445752].

Are there software-only tweaks for quieter measurements?

Reducing gain in driver settings and keeping inputs un-rooted lowers hiss by roughly 6 dB, but cannot match hardware bypass [USB audio cards as measuring instruments].

What is an edge case where the mod gives no benefit?

When recording high-impedance microphones that need pre-amp gain, bypass mode offers no amplification and yields low recorded level.
Generated by the language model.
ADVERTISEMENT