I just bought an H5 to simultaneously record my voice (fed through a Rode wireless go) and engine sound (through an external XLR mic close to the exhaust) on my motorcycle (so pretty loud environment). I was in the 2-3 gain level range and could not figure out why the audio signal was clipping with the levels well in range and clipping indicator not coming on during my first test ride yesterday. Took me a while to find your video but I am glad I did! Thank you for pointing it out and providing a solution - let's hope this works for me. Vielen Dank & Schönen 4. Advent!
Can confirm the problem on the Zoom H6 too! Struggled with it for years... then found out that using the -20db switch kinda reduced the problem, but would have never imagined it could have been because of the gain being TOO LOW! Figures that I always keep it between 1 and 3 😅 It distorts only when input sound it too high, otherwise it does not distort... so weird! Thank you a bunch!!!
I think some of the criticism of the H5 in this discussion is a little off the mark, in the sense that part of what you're running into is a natural consequence of the architecture of the device. The specifications for the H5 state that the maximum input level for inputs 1 & 2, with the -20dB pad engaged, is +22dBu. Now, already things are starting to get messy, because the whole question of what constitutes a "line level" signal is fraught with confusion and multiple standards, but trying to tiptoe carefully around that quicksand pool I'll just say that one version of "line level" equates 0dBVU, or nominally the maximum undistorted level, with +4dBu. The reference level for dBu is 1mW into 600 Ohms, or 774.6mVrms. (And woe betide you if you lose track of whether you're talking about Vpp or Vrms in all of this, because all of a sudden none of the numbers will make sense.) So +4dBu is about 1.228Vrms, and the H5 max input spec of +22dBu is 9.75Vrms (or 27.58Vpp, which is more convenient if you're looking at an oscilloscope). On my H5, with the -20dB pad in, at an input voltage of 27.5Vpp, I hit 0dBfs (in other words, maximum digital signal level) with the gain knob turned to about 3. This means (and here's the punch line) that if you have to turn the gain down below 3 because your input level is too high, then you are exceeding the maximum input level spec of the device. Now before people start screaming that +22dBu is too low, remember that the nominal 0dBVU for this "line level" standard is +4dBu. So that's 18dB of headroom before the front end starts to clip, which seems fairly respectable to me. Now for the architecture part. People are upset because they want the VU meters to tell them if they're overdriving the input stages, but the VU meters can't do that because they are implemented digitally downstream of the ADC. Think about it this way: there is no way, downstream of the ADC, to know if the clipped waveform you are seeing is the result of clipping in the front end, or if that's actually the waveform that was presented to the inputs. A caveat to this argument is that I'm assuming the gain controls are analog attenuators between the input amps and the ADCs. Everything I've seen points to this interpretation, but I haven't actually taken mine apart and reverse engineered it. So, armed with the above knowledge, it should be clear that if you're hitting 0dBfs with the gain set to 3, the solution is not to turn the knob down further, but to reduce the input level somehow. If you can't do it by adjusting whatever the signal is coming from, then inserting a passive in-line attenuator should solve the problem. Does this mean that levels below 3 on the gain knob shouldn't exist? No! There are perfectly legitimate reasons why you might want to turn the level that far down on a channel even if it isn't pushing the maximum input limit. Remember, it's not turning the level down below 3 that's causing the distortion, it's simply that attenuating after the input stage can't do anything about overdriving the input.
Amazing information - I just watched a video of yours on why 32-bit float isn't all that it's cracked up to be, but the better ADCs and the companies designing the equipment such that you *cannot* change the gain means they're probably trying to figure some of this out with the newer recorders. Thanks for pointing out that the file format isn't the only reason why the newer ones "cannot clip" (Which they still can) As someone who never remembers to check his gear before going out, I really appreciate the "set and forget", because if I had to enable the -20db pad in some situation, I can guarantee that I'll forget that until a few gigs later, after wondering why my last few recordings needed so much ampilfication.
Hi Julian. That's a real serious problem. I think you've been very generous in not calling this what it really is, a stock fault with the Zoom. For the price of these things you'd expect a totally analogue gain stage and no digital attenuating/gaining what so ever. I've seen worse on other devices where they have a high bit depth A/D buffer, then everything thereafter is attenuated/gained within the digital domain before final render. This is usually done to apply all the dynamics processing digitally as it's cheaper than doing it in the analogue preamp stage. But the obvious problem is that any issues in the analogue stage will carry through in the digital domain, as you've mentioned. I wouldn't be surprised if on some of these recorders that the limiter is applied in the digital domain, which would also explain why some of them distort. I sometimes go back to tape and a small low noise console just to remind myself of how things used to be. Although you can't beat digital recording as long as all your inputs and outputs are done completely analogue with good analogue components and high resolution A/Ds and D/As inbetween. Cheers, Dave.
You are right, the 0 to 3 gain range should not exist on the H5. And btw you did guess correctly, the H5 also has a built in limiter which is only applied in the digital domain and thus will not prevent any clipping at all...
Wow, das hat mir echt geholfen, ich musst meinen H5 jetzt schon 2 mal zurück schicken, weil ich dachte, es sei ein Defekt, wenn ich im unteren Gain bin, dann kommt so ein, naja, LaserEffekt artiges Geräusch... Aber dann darf ich halt kein niederes Gain mehr nutzen, DANKE!!!!
thanks so much, I had that issue in the H6 and I couldn't find the issue as it wasn't clipping on screen. I tried to change settings as you do and it worked perfectly!
Recorded my first live show recently and this happened. If I would've known about the pad this would not have happened. BUT NOW I DO! Thanks so much glad it wasn't a paid gig.
Very helpful! I'm watching this a day before my first use of the H5! I've tested on myself for standard speach, which has gone well, but this is very handy to know!! Thank you!
Thanks for your video and your findings. There's also a serious problem with the included mic. capsule XYH-5. Sound signals with lots of high frequency content already do clip at a level meter reading of just a little above -12dBFS, and consequently you have audible distortions on the recording. The clipping indicators are flashing, although the meter reading is way below 0dBFS! The XLR inputs don't seem to be plagued by this problem. Test conditions: Create a sine chirp from 20 to 40kHz at 0.8% of full scale in Audacity. (Sampling rate, 96kHz) Record it through the XYH-5 microphone capsule via playback on your speakers, sampling rates 44.1kHz, 48kHz, or 96kHz, stereofile rec. mode. Set the gain controls for meter readings savely below 0dBFS (e.g., -3, -6, or -10dBFS) at all frequencies, i.e. throughout the chirp from low to high. When the chirp reaches high frequencies, (approx. 13-15kHz or so, and higher), the overload lights start flickering, and the audio becomes distorted. (Although meter readings on the H5 still savely below 0dBFS.) Also: Record directly from your interface line out (-10dBV) into the combined mic/line input of the XYH-5 capsule: Same results. Set the gain control so that the level meter reads just below -12dBFS on the H5, no overload indicator lights flashing, no audible distortions.
Thanks Julian. My H6 still clips between 3 and 10 with the xy and stereo capsulse via line out. I get better results recording via the headphone out to my camera but still not great sounding. The onboard recording is outstanding though. I might do a factory reset or look at firmware. Thanks man.
Had a similar problem trying to take the output from a DBX 286s into a Scarlett 2i2. Even turning down the 2i2 gain all the way it would still clip unless the 286s gain was turned down so far that the meter LEDs on the 286s were essentially useless. Consumer level audio devices like the Zoom H range seem to have a remarkable number of problems dealing with any kind of line level signals. Makes it a real pain for those of us who don't understand all the different signal levels before we start buying equipment. Brilliant analysis of the problem and solution as usual, Julian.
Hey, did you feed the signal from the dbx into the XLR input on your 2i2? Because if you go straight into the TRS line level input it should not clip on your 2i2. Cheers!
Julian Krause - I tried multiple different kinds of cables. XLR and both TRS balanced and unbalanced. Nothing really seemed to work like I expected or wanted regardless of how the 2i2 was set (mic vs instrument). Like I said I made it work without clipping but only by drastically reducing the output gain from the DBX 286s. It’s possible there’s something wrong with my unit. Got it used from eBay.
Thx for that usefull information. It turns out that other Zoom Handhelds like the H1 have the same issue! I had that with my last DJ-Set on the H1. Evrything looked fine with no clip-LED blinking, but it sounded like clipped rubish...thx again :)
Thanks for this video! Have you experienced any hissing when using the L/R input with the XY mic capsule? How is your noise floor with that? I just bought an H5 and when using the XY mic, I'm getting very bad hissing. Once I normalize the audio levels, the hissing gets even louder of course and it makes the audio distracting and hard to use. Wondering if this is an issue with my unit.
Damn it Zoom! I knew I heard something wrong outputting from my DBX 286s into the H5. I wanted to avoid engaging the pad to avoid the noise floor completely. Just might have to now.
Funny, I encountered this clipping problem because I hooked up the H5 to a DBX 286s as well! Don't worry about the noise floor too much, when you activate the PAD and feed in a strong signal like the line level from the 286s the noise is going to be miniscule.
Using the pad will actually lower your noise floor, assuming the DBX unit has a better gain stage, which is highly likely. Using the pad will mean that the units preamps aren't being worked as much and the gaining is being taken care of by a better preamp. In this instance the recorder is basically being used as a recorder.
My zoom h5 has a weird static noise That kinda Sounds like water dropping when i turn the gain Above 3. Do you know What causes this or do you know how i can fix it?
thanks for the tip, i also have an issue, that on my recordings one line is always 10 db louder than other, and its not because the input is louder on one channel, this happens when im recording audio output / rec out from DJ mixer, like A&H, the mixer is not showing one channel louder than other, but the recording always ends up with one channel 10db+..... does anyone has similar issue?
I know this video is about 3 years old, but I just tried to use the H5 as Audio Interface in loop back mode which worked out fine .....but ...the quality of the recorded loop back signal is much worse than the original played back signal .... frequency spectrum looks like much attenuation in the high frequencies ..... I checked even that the played back audio has the same resolution/format - not to blame some internal conversion (e.g. 48 to 44 kHz or 24 to 16 bit, ect) .... anybody else found that out ? any work around FW Update ect ? .... It's quite sad, because just to produce a program in high quality audio the recorded audio is a bit flat compared to the original .....
I’d be interested to see you review the zoom h6 as just an audio interface to see how it compares to other audio interfaces performance-wise. This isn’t something you’ve done is it?
I have the zoom h1n. Where do I find the pad -20 db menu? I can't access it. I am trying to run. My dj mixer into the line in input on the zoom but I still get clipping. I cant find the menu screen. Help!🙏🏼 great video by the way!
The same thing happens with the Zoom H1. Once you go below the 5.5 mark, the level is reduced digitally after the input, which actually means if it’s overdriven on the way in and then reduced afterwards. The distorted signal just isn’t as loud but still distorted. The digital circuits can be heard coming in by listening to the hiss level which is the point where they start to work, but unfortunately, after the signal has already been distorted on the way in!!
Alright Ian. You should just use a camera with better audio :) Sorry, I just couldn't resist that. This is another reason why I wouldn't do synced audio, most of these low end recorder-preamp devices simply don't give enough of a benefit over a good camcorder's audio system. Cheers, Dave.
Hi Dave. The problem is only there with extremely loud sources. Live pop groups levels. Generally, it’s not a big deal. If only I could afford a camera with a good preamp .......
Hi Julian ! I was just wondering what do you refer to when you say ‘loud sounds’ a blender machine? A hair dryer? What would be the Db threshold when putting the gain level at 3 and the 20db automatic down grade? Asking you this because I am plannig to Use the H5 to record some small planes single engine sounds. Is the H5 appropriate? Or shoukd I change to a different equipment? Thanks a lot ! And great video !
Great tips! By the way, do you think there's something wrong with my H5? When using XLR inputs and set gain level knob to level 5 and up, I get a lot of hiss. This is for any microphone that I use. Thanks!
I keep getting a clicking noise if it reaches half way (-12) & it also gets a static noise (ground loop.) I have the same issue with my speakers & have a hum destroyer HD400 for the problem. Just waiting for the right cables to arrive. With the H5 connected to the Hum Destroyer, this should deal with any Ground Looping shouldn't it?!
Hi Julian! I love your videos on quircks in audio devices. They are so detailed and in-depth. I have a question. I intend to buy Zoom F1 to use it with the Zoom capsules that I already own as ultra-portable music recording setup with Sony RX0 II (or Z CAM E1 if I lay my hands on that one). I actually own also Zoom ECM-3 cable that would allow me to mount only the capsules, not the device. I need both internal recording to fine-tune sound and audio output to camera to make sync easier or for instant upload. But when I started to think about it I found out a problem. F1 doesn't have line out. Only 2x11mW headphone out. So I started to wonder if this will be enough to feed to camera. Sony seems to have hot input. When I custom made 5-pin XLR to 3.5mm minijack cable (with -12dB resistors removed) for my Rode NT4 it output signal so strong that I can set gain in camera at 1 (out of 30 something). But Z cam E1 has line input. Also Panasonic GH5S that I plan to purchase has line input which would be nice to benefit from. But does Zoom F1 provide strong enough signal? Maybe I should go for something like second-hand Zoom H5 which actually have line out?
Hey, the headphone output should be able to provide a decent level into your camera even though it is not strictly a line level output. So I think the F1 could work for your case.
@@JulianKrause Then what is stronger? Line level or headphone level? I guess the former but buy how much and how to predict that based on output specs?
Hello Julian, I have a Zoom H6. Yesterday i connected my mic to H6 and started recording. When i listened the recording i realized that the signal was low. It kept going like that for 1 min. Than a "click" sound heard and audio boosted around 20db ( which it normally should be) how can i make sure that where the problem is ?
Just wasted 2 hours recording a DJ mix and the hire fee of the studio to have a wonderfully distorted unusable recording. I don’t understand why they don’t let you see the clipping on the meter.
Hello! I have the H5 and i have problems with the input signal is high even when i have the attenuator set to -20db. Am i still getting a clipped signal if i record at 2 even with the -20db? How do you sugges that i can lower the input signal even more? IMHO there should be an -30 db option also..
Hi, I have a Zoom H5. Want to record my trumpet playing. Mostly solo, but also som jazz with Contrabass. Planning to use Shure SM57. What format and adjustment should i use to get optimal quality? Thanks!
Julian, I was thinking about this today with regard to very loud sounds. Many microphones have a -10dB pad so for really loud stuff, I wonder whether this is enough to stop clipping happening? In which case, it might have been handy to have had one on the way in to the recorder before any digital reduction. I get similar with regard to gain levels on the Tascam DR05 which needs a pretty small signal from my mixing desk if connected directly as a line in. It doesn’t clip but has to be set very low, with the desk output also lowered quite a lot.
Hey, a -10dB PAD might help. But as I showed in the video you can always activate a -20dB PAD on the H5. This attenuates the signal before it is A/D converted. So you dont really need a PAD on the mic. Regarding your DR05. When you feed in a strong signal and you have to lower its gain, that's a good thing. This way you will have a good signal-to-noise ratio. If you don't experience any clipping you are fine.
Yes, you’re absolutely right.... the noise floor drops away into virtually nothing as long as a good mic is on the other end. Sorry, I missed the 20dB pad on your video. I thought there might well be one in order to avoid this problem but wasn’t certain.
@@iancraig All good, I also have to confess I was very skeptical about the PAD on the H5 because I was not sure if it is attenuating the signal before or after A/D conversion. That's why I tested it and it turns out to be an analog attenuation which I would not have expected.
Yes Julian. I looked for the manual and when I saw it, I assumed that it was digital and maybe in the wrong place as well. It’s a nice recorder and I battled with h4n or H5 but settled with the H4n in the end. I attenuate from an external mic if that problem appears, but tbh, it’s never been a real issue for me because I mostly record speech and ambience.
Hi Julian - I record our vocal choral concerts on an H5 using the WAV format. For rehearsals, I will record in MP3 format to save space in both storage and transfer. I use the 64K format of MP3. It sounds terrible, with a pulsing distortion. I tolerate it for the other reasons mentioned. Levels never get beyond about -6. Should it sound so terrible? BTW using the XYH-6 capsule mic.
Hey, mp3 is a lossy format. The quality of the mp3 recording is influenced by the bitrate. 64k is a low bitrate and will yield in a bad sound quality. I would suggest you use 256k or 320k. Of course the file size will increase a bit, but the sound quality is way better. Also check if the limiter is turned on. If that's the case, turn it off. The limiter in the H5 is digital and does not prevent clipping. And lastly, try to get your H5 close to the performers. Hope that helps!
@@JulianKrause Thanks for the reply. I know MP3 is a lossy format - of course we listen to most of our tunes in mp3, which certainly don't sound bad. I moved the mic closer to the performers, which didn't seem to help. I can use higher bit rates, which I probably will - I just don't think that even at 64k I should have this strange noise pattern - the other thing, is it seems it varies by frequency. Really appreciate your thoughts!
Hallo Julian, bin gelernter (und daher schon etwas älterer ;-) Radio- u. Fernsehtechniker, mein Englisch ist nicht so gut, daher schreibe ich mal auf Deutsch. Ich mache beruflich sehr viele Sprachaufnahmen mit dem Tascam DR-05V2, wo ich auch oft eine für mich nicht nachvollziehbare Verzerrung auf den Stimmen habe, obwohl die Pegel stimmen. Ich arbeite meist mit Funkstrecken und Headsets, die dann über Line aufgenommen werden, der Rec-Level ist meist im unteren Bereich (zwischen 10 und 14 nach Tascam-Scala). Nachdem ich Dein obiges Video sah, frage ich mich nun: kann es sein, dass es den gleichen Effekt auch beim DR-05 gibt? Vielleicht verwenden die die gleichen A/D-Wandler? Hast Du das mal getestet?
Hi Michael, den DR-05 V2 habe ich selbst noch nicht getestet, aber es kann sehr gut sein, dass es dort ein ähnliches Verhalten gibt. Eventuell ist das Signal von der Funkstrecke zu stark. Was du mal probieren könntest, ist das Ausgangssignal vom Empfänger etwas abzuschwächen und dafür mit einer leicht höheren Verstärkung am DR-05 V2 aufzunehmen. Das sollte das Grundrauschen nicht wirklich beeinflussen, aber eventuell stößt das Signal dann nicht mehr an die Clipping-Grenze. Eine zweite Möglichkeit könnte auch sein, dass das Mikrofonsignal schon in der Funkstrecke verzerrt. Eventuell ist die Mikrofon-Verstärkung am Sender zu hoch eingestellt, sodass das Signal schon im Sender clippt. Dann bekommen alle Geräte in der weiteren Signalkette natürlich nur noch das geclippte Signal mit.
I have the same question. Using the default XYH-5 capsule, if I risk signal clipping below a certain level as for XLR/TRS inputs, which is that value and can I prevent it by activating the -12dB backup recording function (just like the PAD solve the problem with XLR/TRS inputs)? Or does the -12dB attenuation engage after A/D conversion, so it is useless?
@@JulianKrause Hey Julian so came across a problem like this with Sony Cameras Mic Input and Gain settings possible to do some test on this Audio Problem? Seems like a similar Issue to Zoom H5 UA-cam Video Title " Sony A7sIII Audio Distortion Problem SOLVED!! (A7iii as well) " And possibly other Sony Cameras Released around this Period with same pre-amp audio circuit. "The Sony A7iii and A7Siii have a peculiar audio distortion problem if you practice traditional gain staging methods. It seems that having the mic level set to the lowest +1 will actually cause your audio to distort and clip at -9dB when you have plenty of headroom?! " - TLDR Filmmaker
Hey, I don't think the H5 works particularily well as an interface. You always have to turn on the H5 when you start the PC and tell it to connect to your computer and it has it is difficult to connect and control headphones and studio monitors. In a pinch it can work quite decently but if you want something permanently attached to your PC, get a proper interface. Cheers!
@@JulianKrause Thanks for the response! I wasn't really considering buying an H5 because I have a very limited budget. Instead I'm planning to get an UMC202HD because it's the cheapest 2 input interface I could find and I've seen that it's quite decent based on your measurements.
Ich hab jetzt grad keine Lust in englisch zu Schreiben. ;-) Informatives Video! Wenn ich mir das anhand des Zoom H5 genauer überlege, dürfte mein Zoom H6 genauso von dieser "Eigenart" betroffen sein...
I think it is a nice little recorder. I have owned a Zoom H5 for years and the majority of audio on my channel is recorded with the H5. The price-performance ratio is pretty high.
Why is it, that so many Reviews in UA-cam use the Zoom H5, to review devices like the Fethead or Cloudlifter? Was it’s noise on high gains so much worse than other devices, if even rather cheap audio interfaces, like the Behringer U-Phoria ones have a supposedly low noise floor? And if so: Why does everybody buy them? What gives?
Weiß ich ehrlich gesagt nicht. Habe beide Geräte noch nicht getestet. Wenn ich schätzen müsste, würde ich sagen dass der H6 das gleich Problem hat, weil die Vorverstärker Stufe ähnlich aufgebaut ist. Keine Ahnung wie es beim H4n Pro ist.
@@JulianKrause Danke ... und ist das beim L/R Eingang (verbaute Mikrofone und/oder Line/Mic-In 3,5") auch so oder betrifft es nur die Eingangskanäle 1 & 2 ??
Got the H5 since late 2014 and I am still learning how to use it properly. LOL!
I just bought an H5 to simultaneously record my voice (fed through a Rode wireless go) and engine sound (through an external XLR mic close to the exhaust) on my motorcycle (so pretty loud environment). I was in the 2-3 gain level range and could not figure out why the audio signal was clipping with the levels well in range and clipping indicator not coming on during my first test ride yesterday. Took me a while to find your video but I am glad I did! Thank you for pointing it out and providing a solution - let's hope this works for me. Vielen Dank & Schönen 4. Advent!
Can confirm the problem on the Zoom H6 too!
Struggled with it for years... then found out that using the -20db switch kinda reduced the problem, but would have never imagined it could have been because of the gain being TOO LOW! Figures that I always keep it between 1 and 3 😅
It distorts only when input sound it too high, otherwise it does not distort... so weird!
Thank you a bunch!!!
I think some of the criticism of the H5 in this discussion is a little off the mark, in the sense that part of what you're running into is a natural consequence of the architecture of the device. The specifications for the H5 state that the maximum input level for inputs 1 & 2, with the -20dB pad engaged, is +22dBu. Now, already things are starting to get messy, because the whole question of what constitutes a "line level" signal is fraught with confusion and multiple standards, but trying to tiptoe carefully around that quicksand pool I'll just say that one version of "line level" equates 0dBVU, or nominally the maximum undistorted level, with +4dBu. The reference level for dBu is 1mW into 600 Ohms, or 774.6mVrms. (And woe betide you if you lose track of whether you're talking about Vpp or Vrms in all of this, because all of a sudden none of the numbers will make sense.) So +4dBu is about 1.228Vrms, and the H5 max input spec of +22dBu is 9.75Vrms (or 27.58Vpp, which is more convenient if you're looking at an oscilloscope).
On my H5, with the -20dB pad in, at an input voltage of 27.5Vpp, I hit 0dBfs (in other words, maximum digital signal level) with the gain knob turned to about 3. This means (and here's the punch line) that if you have to turn the gain down below 3 because your input level is too high, then you are exceeding the maximum input level spec of the device. Now before people start screaming that +22dBu is too low, remember that the nominal 0dBVU for this "line level" standard is +4dBu. So that's 18dB of headroom before the front end starts to clip, which seems fairly respectable to me.
Now for the architecture part. People are upset because they want the VU meters to tell them if they're overdriving the input stages, but the VU meters can't do that because they are implemented digitally downstream of the ADC. Think about it this way: there is no way, downstream of the ADC, to know if the clipped waveform you are seeing is the result of clipping in the front end, or if that's actually the waveform that was presented to the inputs. A caveat to this argument is that I'm assuming the gain controls are analog attenuators between the input amps and the ADCs. Everything I've seen points to this interpretation, but I haven't actually taken mine apart and reverse engineered it.
So, armed with the above knowledge, it should be clear that if you're hitting 0dBfs with the gain set to 3, the solution is not to turn the knob down further, but to reduce the input level somehow. If you can't do it by adjusting whatever the signal is coming from, then inserting a passive in-line attenuator should solve the problem. Does this mean that levels below 3 on the gain knob shouldn't exist? No! There are perfectly legitimate reasons why you might want to turn the level that far down on a channel even if it isn't pushing the maximum input limit. Remember, it's not turning the level down below 3 that's causing the distortion, it's simply that attenuating after the input stage can't do anything about overdriving the input.
Amazing information - I just watched a video of yours on why 32-bit float isn't all that it's cracked up to be, but the better ADCs and the companies designing the equipment such that you *cannot* change the gain means they're probably trying to figure some of this out with the newer recorders. Thanks for pointing out that the file format isn't the only reason why the newer ones "cannot clip" (Which they still can)
As someone who never remembers to check his gear before going out, I really appreciate the "set and forget", because if I had to enable the -20db pad in some situation, I can guarantee that I'll forget that until a few gigs later, after wondering why my last few recordings needed so much ampilfication.
Hi Julian. That's a real serious problem. I think you've been very generous in not calling this what it really is, a stock fault with the Zoom. For the price of these things you'd expect a totally analogue gain stage and no digital attenuating/gaining what so ever. I've seen worse on other devices where they have a high bit depth A/D buffer, then everything thereafter is attenuated/gained within the digital domain before final render. This is usually done to apply all the dynamics processing digitally as it's cheaper than doing it in the analogue preamp stage. But the obvious problem is that any issues in the analogue stage will carry through in the digital domain, as you've mentioned. I wouldn't be surprised if on some of these recorders that the limiter is applied in the digital domain, which would also explain why some of them distort. I sometimes go back to tape and a small low noise console just to remind myself of how things used to be. Although you can't beat digital recording as long as all your inputs and outputs are done completely analogue with good analogue components and high resolution A/Ds and D/As inbetween. Cheers, Dave.
You are right, the 0 to 3 gain range should not exist on the H5. And btw you did guess correctly, the H5 also has a built in limiter which is only applied in the digital domain and thus will not prevent any clipping at all...
Wow, das hat mir echt geholfen, ich musst meinen H5 jetzt schon 2 mal zurück schicken, weil ich dachte, es sei ein Defekt, wenn ich im unteren Gain bin, dann kommt so ein, naja, LaserEffekt artiges Geräusch...
Aber dann darf ich halt kein niederes Gain mehr nutzen, DANKE!!!!
thanks so much, I had that issue in the H6 and I couldn't find the issue as it wasn't clipping on screen. I tried to change settings as you do and it worked perfectly!
Recorded my first live show recently and this happened. If I would've known about the pad this would not have happened. BUT NOW I DO! Thanks so much glad it wasn't a paid gig.
Very helpful! I'm watching this a day before my first use of the H5! I've tested on myself for standard speach, which has gone well, but this is very handy to know!! Thank you!
Thanks for your video and your findings.
There's also a serious problem with the included mic. capsule XYH-5.
Sound signals with lots of high frequency content already do clip at a level meter reading of just a little above -12dBFS, and consequently you have audible distortions on the recording. The clipping indicators are flashing, although the meter reading is way below 0dBFS!
The XLR inputs don't seem to be plagued by this problem.
Test conditions:
Create a sine chirp from 20 to 40kHz at 0.8% of full scale in Audacity. (Sampling rate, 96kHz) Record it through the XYH-5 microphone capsule via playback on your speakers, sampling rates 44.1kHz, 48kHz, or 96kHz, stereofile rec. mode. Set the gain controls for meter readings savely below 0dBFS (e.g., -3, -6, or -10dBFS) at all frequencies, i.e. throughout the chirp from low to high. When the chirp reaches high frequencies, (approx. 13-15kHz or so, and higher), the overload lights start flickering, and the audio becomes distorted. (Although meter readings on the H5 still savely below 0dBFS.)
Also: Record directly from your interface line out (-10dBV) into the combined mic/line input of the XYH-5 capsule: Same results.
Set the gain control so that the level meter reads just below -12dBFS on the H5, no overload indicator lights flashing, no audible distortions.
Thanks Julian. My H6 still clips between 3 and 10 with the xy and stereo capsulse via line out. I get better results recording via the headphone out to my camera but still not great sounding. The onboard recording is outstanding though. I might do a factory reset or look at firmware. Thanks man.
Had a similar problem trying to take the output from a DBX 286s into a Scarlett 2i2. Even turning down the 2i2 gain all the way it would still clip unless the 286s gain was turned down so far that the meter LEDs on the 286s were essentially useless. Consumer level audio devices like the Zoom H range seem to have a remarkable number of problems dealing with any kind of line level signals. Makes it a real pain for those of us who don't understand all the different signal levels before we start buying equipment. Brilliant analysis of the problem and solution as usual, Julian.
Hey, did you feed the signal from the dbx into the XLR input on your 2i2? Because if you go straight into the TRS line level input it should not clip on your 2i2. Cheers!
Julian Krause - I tried multiple different kinds of cables. XLR and both TRS balanced and unbalanced. Nothing really seemed to work like I expected or wanted regardless of how the 2i2 was set (mic vs instrument). Like I said I made it work without clipping but only by drastically reducing the output gain from the DBX 286s. It’s possible there’s something wrong with my unit. Got it used from eBay.
Great tip Julian, the -20dB pad is definitely the answer. I appreciate it.
Thanks very much! Was having this problem with levels set at about 2.5. Will try your solution at our next band rehearsal. Very Helpful!
That's exactly what happened to me, recorded below 3 at a very low level but the audio resulted unusable, thanks for the info!
Thx for that usefull information. It turns out that other Zoom Handhelds like the H1 have the same issue! I had that with my last DJ-Set on the H1. Evrything looked fine with no clip-LED blinking, but it sounded like clipped rubish...thx again :)
I didn't know mics clipped at 0 decibals too... crazy. Thanks for the help. :)
High quality information and presentation. Thank you.
Thank you for your tests and explanation. I am so pissed at Zoom, so many recordings of mine are unusable because of this
Same problem with Q2n, super helpful, thanks.
Julien, you have THE best channel out here! Love it! Would you say I could do well with the H5 for VO? I have the Babyface Pro as well. Thanks!
Thank you so much! Been having this problem for a while
Thanks for this video! Have you experienced any hissing when using the L/R input with the XY mic capsule? How is your noise floor with that?
I just bought an H5 and when using the XY mic, I'm getting very bad hissing. Once I normalize the audio levels, the hissing gets even louder of course and it makes the audio distracting and hard to use. Wondering if this is an issue with my unit.
Thanks for this! Good to know where the weeknesses are.
This is fantastic, danke schoen!
Damn it Zoom! I knew I heard something wrong outputting from my DBX 286s into the H5. I wanted to avoid engaging the pad to avoid the noise floor completely. Just might have to now.
Funny, I encountered this clipping problem because I hooked up the H5 to a DBX 286s as well! Don't worry about the noise floor too much, when you activate the PAD and feed in a strong signal like the line level from the 286s the noise is going to be miniscule.
Using the pad will actually lower your noise floor, assuming the DBX unit has a better gain stage, which is highly likely. Using the pad will mean that the units preamps aren't being worked as much and the gaining is being taken care of by a better preamp. In this instance the recorder is basically being used as a recorder.
David Harry 😎 thanks
My zoom h5 has a weird static noise That kinda Sounds like water dropping when i turn the gain Above 3. Do you know What causes this or do you know how i can fix it?
Thanks for this work. I wonder if the H6 has the same default ?? Thx again.
Interesting to note. Thanks for the heads up👍
thanks for the tip, i also have an issue, that on my recordings one line is always 10 db louder than other, and its not because the input is louder on one channel, this happens when im recording audio output / rec out from DJ mixer, like A&H, the mixer is not showing one channel louder than other, but the recording always ends up with one channel 10db+..... does anyone has similar issue?
I know this video is about 3 years old, but I just tried to use the H5 as Audio Interface in loop back mode which worked out fine .....but ...the quality of the recorded loop back signal is much worse than the original played back signal .... frequency spectrum looks like much attenuation in the high frequencies ..... I checked even that the played back audio has the same resolution/format - not to blame some internal conversion (e.g. 48 to 44 kHz or 24 to 16 bit, ect) .... anybody else found that out ? any work around FW Update ect ? .... It's quite sad, because just to produce a program in high quality audio the recorded audio is a bit flat compared to the original .....
Does this work in the left and right channels as well? It looks like its only a setting for external mics
I’d be interested to see you review the zoom h6 as just an audio interface to see how it compares to other audio interfaces performance-wise. This isn’t something you’ve done is it?
Zoom actually acknowledge the problem but they don't go into details. Just state that at volumes below 3 the noise is present. Maybe clipping too.
I have the zoom h1n. Where do I find the pad -20 db menu? I can't access it. I am trying to run. My dj mixer into the line in input on the zoom but I still get clipping. I cant find the menu screen. Help!🙏🏼 great video by the way!
Good review.
The same thing happens with the Zoom H1. Once you go below the 5.5 mark, the level is reduced digitally after the input, which actually means if it’s overdriven on the way in and then reduced afterwards. The distorted signal just isn’t as loud but still distorted. The digital circuits can be heard coming in by listening to the hiss level which is the point where they start to work, but unfortunately, after the signal has already been distorted on the way in!!
Good point, I made a video about that a while back: ua-cam.com/video/Dv663l84Vlo/v-deo.html
Alright Ian. You should just use a camera with better audio :) Sorry, I just couldn't resist that. This is another reason why I wouldn't do synced audio, most of these low end recorder-preamp devices simply don't give enough of a benefit over a good camcorder's audio system. Cheers, Dave.
Hi Dave. The problem is only there with extremely loud sources. Live pop groups levels. Generally, it’s not a big deal.
If only I could afford a camera with a good preamp .......
this problem solved in frimeware update 2.10 from zoom
Very informative.. but this guy never blinks
Mr Robot was modelled after him.
Thanks man you saved my life !!!!!!!!!!!!
im getting kind of flutter sound , wobbly when used as audio interface . any idea why ?
Hi Julian ! I was just wondering what do you refer to when you say ‘loud sounds’ a blender machine? A hair dryer? What would be the Db threshold when putting the gain level at 3 and the 20db automatic down grade? Asking you this because I am plannig to Use the H5 to record some small planes single engine sounds. Is the H5 appropriate? Or shoukd I change to a different equipment?
Thanks a lot ! And great video !
Thanks
Interesting! Thanks for the tips.
thank you, absolute legend!!! :)
Great tips! By the way, do you think there's something wrong with my H5? When using XLR inputs and set gain level knob to level 5 and up, I get a lot of hiss. This is for any microphone that I use. Thanks!
What mic are you using? The preamps of the H5 are not the cleanest but it should't be crazy high noise.
Thanks for sharing.
I keep getting a clicking noise if it reaches half way (-12) & it also gets a static noise (ground loop.) I have the same issue with my speakers & have a hum destroyer HD400 for the problem. Just waiting for the right cables to arrive. With the H5 connected to the Hum Destroyer, this should deal with any Ground Looping shouldn't it?!
Great video Julian! Thanks a lot!
Is this happening to a lot of people? Does this happen with the h6?
merci pour cette astuce !!
I'm going to check this out but I'm getting clipping at lvl 4 gain even. Hope this works
Hi Julian! I love your videos on quircks in audio devices. They are so detailed and in-depth. I have a question. I intend to buy Zoom F1 to use it with the Zoom capsules that I already own as ultra-portable music recording setup with Sony RX0 II (or Z CAM E1 if I lay my hands on that one). I actually own also Zoom ECM-3 cable that would allow me to mount only the capsules, not the device. I need both internal recording to fine-tune sound and audio output to camera to make sync easier or for instant upload. But when I started to think about it I found out a problem. F1 doesn't have line out. Only 2x11mW headphone out. So I started to wonder if this will be enough to feed to camera. Sony seems to have hot input. When I custom made 5-pin XLR to 3.5mm minijack cable (with -12dB resistors removed) for my Rode NT4 it output signal so strong that I can set gain in camera at 1 (out of 30 something). But Z cam E1 has line input. Also Panasonic GH5S that I plan to purchase has line input which would be nice to benefit from. But does Zoom F1 provide strong enough signal? Maybe I should go for something like second-hand Zoom H5 which actually have line out?
Hey,
the headphone output should be able to provide a decent level into your camera even though it is not strictly a line level output. So I think the F1 could work for your case.
@@JulianKrause Then what is stronger? Line level or headphone level? I guess the former but buy how much and how to predict that based on output specs?
Hello Julian, I have a Zoom H6. Yesterday i connected my mic to H6 and started recording. When i listened the recording i realized that the signal was low. It kept going like that for 1 min. Than a "click" sound heard and audio boosted around 20db ( which it normally should be) how can i make sure that where the problem is ?
Just wasted 2 hours recording a DJ mix and the hire fee of the studio to have a wonderfully distorted unusable recording.
I don’t understand why they don’t let you see the clipping on the meter.
Great info!
Hello!
I have the H5 and i have problems with the input signal is high even when i have the attenuator set to -20db. Am i still getting a clipped signal if i record at 2 even with the -20db?
How do you sugges that i can lower the input signal even more? IMHO there should be an -30 db option also..
Thank you!
Hi, I have a Zoom H5. Want to record my trumpet playing. Mostly solo, but also som jazz with Contrabass. Planning to use Shure SM57. What format and adjustment should i use to get optimal quality? Thanks!
How do I reduce gain on H4n Pro ?
But when I raise the knob above 6dB I get so much noise....
I found that this only works when recording to SD card, do you know if there is a simillar solution when using it as an audio interface?
Da bin ich ja froh, dass ich einen Tascam DR60 II habe. Bei dem ist nur der Kopfhörerausgang schlecht.
Julian, I was thinking about this today with regard to very loud sounds. Many microphones have a -10dB pad so for really loud stuff, I wonder whether this is enough to stop clipping happening? In which case, it might have been handy to have had one on the way in to the recorder before any digital reduction.
I get similar with regard to gain levels on the Tascam DR05 which needs a pretty small signal from my mixing desk if connected directly as a line in. It doesn’t clip but has to be set very low, with the desk output also lowered quite a lot.
Hey, a -10dB PAD might help. But as I showed in the video you can always activate a -20dB PAD on the H5. This attenuates the signal before it is A/D converted. So you dont really need a PAD on the mic.
Regarding your DR05. When you feed in a strong signal and you have to lower its gain, that's a good thing. This way you will have a good signal-to-noise ratio. If you don't experience any clipping you are fine.
Yes, you’re absolutely right.... the noise floor drops away into virtually nothing as long as a good mic is on the other end. Sorry, I missed the 20dB pad on your video. I thought there might well be one in order to avoid this problem but wasn’t certain.
@@iancraig All good, I also have to confess I was very skeptical about the PAD on the H5 because I was not sure if it is attenuating the signal before or after A/D conversion. That's why I tested it and it turns out to be an analog attenuation which I would not have expected.
Yes Julian. I looked for the manual and when I saw it, I assumed that it was digital and maybe in the wrong place as well. It’s a nice recorder and I battled with h4n or H5 but settled with the H4n in the end. I attenuate from an external mic if that problem appears, but tbh, it’s never been a real issue for me because I mostly record speech and ambience.
Awesome
Hi Julian - I record our vocal choral concerts on an H5 using the WAV format. For rehearsals, I will record in MP3 format to save space in both storage and transfer. I use the 64K format of MP3. It sounds terrible, with a pulsing distortion. I tolerate it for the other reasons mentioned. Levels never get beyond about -6. Should it sound so terrible? BTW using the XYH-6 capsule mic.
Hey, mp3 is a lossy format. The quality of the mp3 recording is influenced by the bitrate. 64k is a low bitrate and will yield in a bad sound quality. I would suggest you use 256k or 320k. Of course the file size will increase a bit, but the sound quality is way better.
Also check if the limiter is turned on. If that's the case, turn it off. The limiter in the H5 is digital and does not prevent clipping.
And lastly, try to get your H5 close to the performers. Hope that helps!
@@JulianKrause Thanks for the reply. I know MP3 is a lossy format - of course we listen to most of our tunes in mp3, which certainly don't sound bad. I moved the mic closer to the performers, which didn't seem to help. I can use higher bit rates, which I probably will - I just don't think that even at 64k I should have this strange noise pattern - the other thing, is it seems it varies by frequency. Really appreciate your thoughts!
Hallo Julian, bin gelernter (und daher schon etwas älterer ;-) Radio- u. Fernsehtechniker, mein Englisch ist nicht so gut, daher schreibe ich mal auf Deutsch. Ich mache beruflich sehr viele Sprachaufnahmen mit dem Tascam DR-05V2, wo ich auch oft eine für mich nicht nachvollziehbare Verzerrung auf den Stimmen habe, obwohl die Pegel stimmen. Ich arbeite meist mit Funkstrecken und Headsets, die dann über Line aufgenommen werden, der Rec-Level ist meist im unteren Bereich (zwischen 10 und 14 nach Tascam-Scala). Nachdem ich Dein obiges Video sah, frage ich mich nun: kann es sein, dass es den gleichen Effekt auch beim DR-05 gibt? Vielleicht verwenden die die gleichen A/D-Wandler? Hast Du das mal getestet?
Hi Michael, den DR-05 V2 habe ich selbst noch nicht getestet, aber es kann sehr gut sein, dass es dort ein ähnliches Verhalten gibt. Eventuell ist das Signal von der Funkstrecke zu stark. Was du mal probieren könntest, ist das Ausgangssignal vom Empfänger etwas abzuschwächen und dafür mit einer leicht höheren Verstärkung am DR-05 V2 aufzunehmen. Das sollte das Grundrauschen nicht wirklich beeinflussen, aber eventuell stößt das Signal dann nicht mehr an die Clipping-Grenze.
Eine zweite Möglichkeit könnte auch sein, dass das Mikrofonsignal schon in der Funkstrecke verzerrt. Eventuell ist die Mikrofon-Verstärkung am Sender zu hoch eingestellt, sodass das Signal schon im Sender clippt. Dann bekommen alle Geräte in der weiteren Signalkette natürlich nur noch das geclippte Signal mit.
@@JulianKrause Danke, das werde ich mal testen, ich habe am Wochenende wieder eine Veranstaltung aufzunehmen!
Thx, as i understand this does not apply to XYH-5 capsule?
I have the same question. Using the default XYH-5 capsule, if I risk signal clipping below a certain level as for XLR/TRS inputs, which is that value and can I prevent it by activating the -12dB backup recording function (just like the PAD solve the problem with XLR/TRS inputs)? Or does the -12dB attenuation engage after A/D conversion, so it is useless?
does mic inputs suffer from this issue?
Great detective work. Can Zoom fix this with a firmware upgrade?
Hard to say, I think this depends on how the gain setting is implemented on the H5. It might be possible.
@@JulianKrause Hey Julian so came across a problem like this with Sony Cameras Mic Input and Gain settings possible to do some test on this Audio Problem? Seems like a similar Issue to Zoom H5
UA-cam Video Title " Sony A7sIII Audio Distortion Problem SOLVED!! (A7iii as well) " And possibly other Sony Cameras Released around this Period with same pre-amp audio circuit.
"The Sony A7iii and A7Siii have a peculiar audio distortion problem if you practice traditional gain staging methods. It seems that having the mic level set to the lowest +1 will actually cause your audio to distort and clip at -9dB when you have plenty of headroom?! " - TLDR Filmmaker
@@JulianKrause Here's the Link to the video ua-cam.com/video/Bh_gfEsFGtA/v-deo.html
very helpful! it's too bad this function exists. Not very intuitive that a lower more conservative gain would backfire. Alas, we live and we learn.
I'm curious about how good it is as an audio interface. Can you make a video about how it performs?
Hey, I don't think the H5 works particularily well as an interface. You always have to turn on the H5 when you start the PC and tell it to connect to your computer and it has it is difficult to connect and control headphones and studio monitors. In a pinch it can work quite decently but if you want something permanently attached to your PC, get a proper interface. Cheers!
@@JulianKrause Thanks for the response! I wasn't really considering buying an H5 because I have a very limited budget. Instead I'm planning to get an UMC202HD because it's the cheapest 2 input interface I could find and I've seen that it's quite decent based on your measurements.
Ich hab jetzt grad keine Lust in englisch zu Schreiben. ;-)
Informatives Video!
Wenn ich mir das anhand des Zoom H5 genauer überlege, dürfte mein Zoom H6 genauso von dieser "Eigenart" betroffen sein...
Habe zwar keinen H6 um das zu testen, würde aber auch schätzen, dass das dort genauso ist. Die Elektronik vom H5 und H6 sind sehr ähnlich.
sir what u think of zoom h5 for films?????
I think it is a nice little recorder. I have owned a Zoom H5 for years and the majority of audio on my channel is recorded with the H5. The price-performance ratio is pretty high.
Why is it, that so many Reviews in UA-cam use the Zoom H5, to review devices like the Fethead or Cloudlifter? Was it’s noise on high gains so much worse than other devices, if even rather cheap audio interfaces, like the Behringer U-Phoria ones have a supposedly low noise floor? And if so: Why does everybody buy them? What gives?
Fuck! Just when I decided to send it back!
Danke. Haben H4nPro und H6 dasselbe Problem?
Weiß ich ehrlich gesagt nicht. Habe beide Geräte noch nicht getestet. Wenn ich schätzen müsste, würde ich sagen dass der H6 das gleich Problem hat, weil die Vorverstärker Stufe ähnlich aufgebaut ist. Keine Ahnung wie es beim H4n Pro ist.
@@JulianKrause Danke ... und ist das beim L/R Eingang (verbaute Mikrofone und/oder Line/Mic-In 3,5") auch so oder betrifft es nur die Eingangskanäle 1 & 2 ??
Mines sticky , I don’t like it
does the H6 have the same problem?
yes it has
@@breakingnews9173 this sucks a lot
please blink.
blink
is this a robot?
01001001 00100111 01101101 00100000 01101110 01101111 01110100 00100000 01100001 00100000 01110010 01101111 01100010 01101111 01110100 00100000 00111010 00101001
lol i know, right?