I have used many Zoom H-series recorders over the years, sometimes pushing them outside their design intentions. This page collects the “secrets” I’ve discovered through testing and experience. It will grow as I learn more. Get in touch if you have anything to add!
- If the power is cut (via USB or battery) then the file is preserved (except for last few seconds)
- The USB input accepts a wide range of voltages with equal performance.
It’s OK to wire a LiPo battery directly to the USB +/- lines, without a boost converter. (Make sure your battery has a protection board, because the recorder will discharge to 2.15V if left unchecked, which will kill your cells.) The recorder won’t show battery levels when powered by USB, so you’ll need to provide something external.
- The battery terminals accept a wide range too, but > 3.3v seems like a bad idea. (LCD backlight gets brighter. Something isn’t regulated!) So don’t use a LiPo!
- The limiter on this recorder adds significant hiss if the gain is high. But, it seems to be a real analog limiter, so it handles very hot signals gracefully, and always prevents clipping in my experience.
- The auto gain control is terrible. It seems to have several distinct levels and wacky timing that won’t sound good on dynamic material. Stick with the limiter instead.
- The plug-in-power is always enabled on the 1/8″ mic input.
- Hand-holding such a tiny recorder can be maddening. I made a 3D-printable shock mount that helps isolate vibration.
- If you’re recording in a quiet place with the internal mics, you’ll notice that the plastic volume rocker rattles around, ruining your recording. Put a piece of tape over it and enjoy the silence.
- Menu settings are retained indefinitely if batteries are removed, but only if the recorder was shut down normally via the switch. If power is lost unexpectedly, any changed settings aren’t saved. (On the next reboot, the previously saved settings are restored. It won’t reset to factory defaults or anything tragic like that.)
- I was testing the recorder as a simple 5V mic preamp, so I wondered how to make it turn on whenever power is applied. (Usually you need to slide the spring-loaded power switch down for a few seconds to turn it on.) If you keep the switch engaged forever (either mechanically or electrically) the recorder will power on whenever it receives voltage from the battery terminals or USB. The UI is accessible while the power switch is held down, BUT any changes you make aren’t restored on the next reboot (because the recorder wasn’t shut down gracefully, as noted earlier).
- The clock resets after 15-20 minutes without power. It will force you to reset the clock on the next boot. The recorder doesn’t pass audio on the clock setting screen, but if you wait a few seconds and withdraw power momentarily it will charge the internal supercap, save the default clock setting and allow a normal boot next time. (If you’re controlling the power programmatically, you should be able to get reliable operation by turning the power on – off – on each time you want to use the recorder.)
Zoom H1n Voltage Testing
|Battery Icon||NiMH mode||Alkaline mode||Lithium mode||USB||(notes)|
|3 bars||> 2.54||> 2.66||> 2.97||n/a|
|2 bars||2.49 – 2.54||2.47 – 2.66||2.88 – 2.97||n/a|
|1 bar||2.34 – 2.49||2.22 – 2.47||2.69 – 2.88||n/a|
|0 bars||2.30 – 2.34||2.15 – 2.22||2.62 – 2.69||n/a||(backlight turns off when voltage falls to this level)|
|Shutdown||< 2.30||< 2.15||< 2.62||< 2.15|
* The above measurements were done during discharge. There is hysteresis so these values would be slightly different for rising voltages:
Zoom H1n Current Draw Testing @ 5V via USB input
|Idle||48K WAV Recording||MP3 320k Recording||+ Backlight|
|70 mA||70 mA||80 mA||add 10 mA|
- If the power is cut (via USB or battery) then the file is preserved (except for last few secs)
- I haven’t done extensive voltage testing, but I can confirm that the USB input supports voltages down to 1.7v before shutdown. (The info about the Zoom H1n also applies here.)
- The lo cut and limiter seem digital and they don’t improve headroom at all. With hot inputs I got much better dynamic range with the cheaper Zoom H1n.
Zoom H5 Current Draw Testing @ 5V via USB input
|Idle – 1/8″ Mic with Plugin Power (44k / 96k)||Idle – XLR pair (44k / 96k)||Idle – XLR pair (44k) P48||Idle – XLR pair (44k) P24||Idle – XLR pair (44k) P12||+ Recording||+ Backlight|
|117 / 125 mA||187 / 197 mA||285 mA||207 mA||193 mA||add 10 mA||add 25 mA|
The Zoom H2n UI is similar to the other H-series recorders, but with additional surround features.
- The H2n possesses something weird: an inexpensive electret mic capsule with a figure-8 / bi-directional pattern. These are as rare as hen’s teeth, so I took one out to experiment with it. The verdict is: You could make a very tiny MS mic with it, but it’s way too noisy for quiet recordings.
- I’ve had problems with the mic selector switch on the top of the recorder. The contacts need to be exercised occasionally.
- I have some 3D-printable mods to make the recorder more convenient.
File Naming and Spanning
- (This info applies to the H1n, but the other recorders are probably the same.)
- If the menu is set to name files according to date (rather than sequential numbers) you’ll get files formatted in this format: YYMMDD-HHMMSS.WAV (like “220222-134308.WAV”)
- If recording overflows the internal 2GB file size limit, it spans to another file seemlessly
- On the Zoom H1n (and others?), the first file in a spanned set is named correctly, but later files have incorrect filenames (and creation dates). If you are processing these files algorithmically, you’ll be in trouble if you depend on filenames having a relationship to their duration and time of day. A real-world example:
|Filename||Duration||Filename should be|
|220314-151617.MP3||14:54:47||(first file in spanned set has correct name)|
Maximum File Limits
Zoom recorders seem to have issues with large numbers of files. Maybe the recorder does some sort of indexing of the filesystem each time a recording is requested? On some recorders this must be so inefficient that a few hundred files is enough to backlog the queue and stop UI services. This problem manifests slightly differently in each model:
- Zoom H2n: As the number of files grows, the recorder will have a growing latency between the moment you press Rec and moment the recording starts. After a few hundred recordings, you’ll notice several seconds of latency. (Re-format the SD card to fix it. Simple power-cycling will not help. I tried several cards and formatting methods.)
- Zoom H5: It records fine until it displays an error after 500 files. If you manually switch it to the next folder, it will continue for another 500, but there’s no automatic way to do that.
- Zoom H1n: Same as H5, but there are no folders so 500 files is a hard max.
- Zoom F1? According to this blog post, the Zoom F1 also has a 500 file maximum: https://hackaday.io/project/183340/logs
Remote Control Protocol
The H2n and H5 can be controlled with a wired controller via their 2.5mm TRRS remote jack. (Yes, TRRS: 4 conductors). I used the info from Hacking the Zoom H2n Remote Control to make a timer remote with a ESP32 and its internal UART. The same protocol works for the H5 as well. (I’ve only used it for basic REC / STOP control, without bidirectional communication.)
Note: I eventually abandoned this project because of the file limitations on the Zoom H2n (mentioned above). After a few hundred files, the whole thing became unresponsive so it didn’t make sense as a timed recorder for gathering data. An AudioMoth would work a lot better (but it’s mono).
Hacking Zoom Mic Modules
I spent a lot of time trying to reverse-engineer the modular mic port for interchangeable capsules (The XYH-5, XYH-6, MSH-6, SSH-6, SGH-6, etc). The capsules contain a stereo mic preamp with gain control and pretty good specs, so I was hoping to use them as a front end for other recorders. I never got it to work, so I offer the info here in case you can take the ball and run with it!
- When I powered the module with 5V I was able to pass audio to the pins identified as L and R out.
- The output seriously lacks bass, even though it seems to have a buffer opamp inside. Is the HPF filter inside the module, controlled by a CV signal from the recorder? (I thought the HPF is digital on this machine?)
- I can’t get it to send the PIP bias voltage to the external mic input. This is controlled in the recorder’s menu so it makes sense that it’s gated somehow. Sadly I couldn’t figure out the mechanism.
- Each mic module signals its identity to the recorder somehow. (The UI changes to reflect the features of each module.) Maybe one of the pins I marked with question marks is attached to a voltage divider that gets read by an ADC in the recorder?