JCALLY JM20 Headphone Dongle Review

[solderdude]

I have the jm20 that I sometimes use with my hd600 and wondering if the max or another dongle would make a difference. When I go louder it almost feels like it's running out of gas but could be placebo. It also warms up more than I expected. I haven't done a side by side with other gear I have.. this is my portable setup.

The JM20 (just like almost all other dongles) is rather limited in output voltage 2V = 13mW in 300ohm meaning the HD600 can only reach 109dB peak SPL = between 90 and 100dB average depending on the music which means moderately loud at best.
So while the sound will stay clean up to the max level it can reach you may still be left wanting, especially when you add 5 dB of bass (which isn't a bad idea for the HD600).
In that case you can deduct that 5dB and will end up with average levels between 85 and 95dB SPL.

 

[bequietjk]

Recently I noticed my dongle getting extremely hot to the touch after 20 minutes or so and the audio seems to be clipping at that heat level. Audio fidelity of course is insanely good and would love to keep it but I would rather spend more on something that won't heat up so much and either spend more or take a hit on audio fidelity. Any recommendations?

 

[staticV3]

Recently I noticed my dongle getting extremely hot to the touch after 20 minutes or so and the audio seems to be clipping at that heat level. Audio fidelity of course is insanely good and would love to keep it but I would rather spend more on something that won't heat up so much and either spend more or take a hit on audio fidelity. Any recommendations?

Apple headphone adapter

Consumes one third as much power as the JM20.

 

[nanook]

Since I seem to have lost my JM20 I'm interested as well.
Do you mean this one?

Review: Apple vs Google USB-C Headphone Adapters

This is a review, detailed measurements and comparison of Apple's USB-C adapter to the current and last version of Google Pixel headphone adapters. The Apple adapter costs just $9 including one day shipping for free. The Google dongle costs $12. Not that any of these are large by any stretch...

www.audiosciencereview.com

Or is there a new revision out in the meantime?

 

[staticV3]

Since I seem to have lost my JM20 I'm interested as well.
Do you mean this one?

Review: Apple vs Google USB-C Headphone Adapters

This is a review, detailed measurements and comparison of Apple's USB-C adapter to the current and last version of Google Pixel headphone adapters. The Apple adapter costs just $9 including one day shipping for free. The Google dongle costs $12. Not that any of these are large by any stretch...

www.audiosciencereview.com

Or is there a new revision out in the meantime?

There is a new revision out, actually as the product code was changed from
MU7E2ZM/A (Europe) / MU7E2AM/A (USA) to
MW2Q3ZM/A (Europe) / MW2Q3AM/A (USA)

Though AFAIK the model number has remained the same at A2155 (Europe) / A2049 (USA).

Whether the revision brought any changes to performance, functionality, or power consumption with it, I don't know.

 

[terminus12]

Hi. You can no longer buy the Jcally JM20 or JM28 on AliExpress. Have they been replaced by something else? The JM20 Max is too powerful and drains the phone's battery; I need 2V output, good specs, and low power consumption—ideally in the same form factor as the JM20.

 

[terryforsythe]

The JM20 Max is too powerful and drains the phone's battery;

I have not noticed that being an issue with my JM20 Max using my Samsung phone. I use it on cross-country flights, and have plenty of charge left when I land. I don't remember how much charge it uses, but after landing I never feel like I need to soon recharge the phone.

 

[fws]

Hi. You can no longer buy the Jcally JM20 or JM28 on AliExpress. Have they been replaced by something else? The JM20 Max is too powerful and drains the phone's battery; I need 2V output, good specs, and low power consumption—ideally in the same form factor as the JM20.

There use to be CS-Pro CS43131 with the same chip config, it seems to be gone as well, but I found one seller with a bit later Hi-Pro CS43198. I don't recall seeing measurements of them, CS-Pro seemed audibly ok (had some weird behavior initially, but changing USB C cable fixed it).

https://aliexpress.com/item/1005008257454242.html

Hiby FC3 was ok according to amirm's measurements, seems to be gone from Ali, but still available in local stores, unfortunately often at quite inflated price. Both are a small "pendrive" sized dongle with a detachable cable though.

Seems like there is a gap in the market now. JM7 Max has 2V if I recall it right, but also the annoying fade-in from build in power saving.

 

[Somafunk]

Hi. You can no longer buy the Jcally JM20 or JM28 on AliExpress. Have they been replaced by something else? The JM20 Max is too powerful and drains the phone's battery; I need 2V output, good specs, and low power consumption—ideally in the same form factor as the JM20.

You could ask crinacle if his new dongle dac has 2v output, I can’t see any specifics on his link below

CrinEar Protocol Micro Ultra-Portable DAC-Amplifier

Inspired by the flagship Protocol Max, the CrinEar Protocol Micro delivers the essential Protocol experience in a streamlined, ultra-affordable package, making high-quality portable audio more accessible than ever. The Protocol Micro is a compact USB-C to 3.5mm DAC-amplifier that brings...

hangout.audio

Or there is this version

Truthear KEYX Portable DAC-Amplifier

Hunting for a Tiny DAC-Amplifier? The Truthear KeyX is an ultra-compact, right-angled portable DAC-amplifier designed for both audiophiles and everyday users who desire a convenient and clean sound source from their USB-C devices. Powered by the Cirrus Logic CS43131 chip, this powerful...

hangout.audio

 

[fws]

You could ask crinacle if his new dongle dac has 2v output, I can’t see any specifics on his link below

CrinEar Protocol Micro Ultra-Portable DAC-Amplifier

Inspired by the flagship Protocol Max, the CrinEar Protocol Micro delivers the essential Protocol experience in a streamlined, ultra-affordable package, making high-quality portable audio more accessible than ever. The Protocol Micro is a compact USB-C to 3.5mm DAC-amplifier that brings...

hangout.audio

Modified SPV4040, that is 1V and I'm not 100% sure if I remember it right, but I think that would be CB1200AU = $3 Hi-Max dongle often recommended here (I have it and it is ok).

 

[terminus12]

Thanks, everyone. I bought the Apple A2049 dongle—I managed to find it in Europe for €11. It drives both my HD560S and HE400SE perfectly. I'm using Oratory's PEQ settings for both. I'm really surprised that 1V gets the job done. I don't need anything else.

 

[CedarX]

Modified SPV4040, that is 1V and I'm not 100% sure if I remember it right, but I think that would be CB1200AU = $3 Hi-Max dongle often recommended here (I have it and it is ok).

Indeed, TTGK Station app identifies it as a CB1200-AU (Vid/Pid 0x3302 / 0xC20F). It is WalkPlay-compatible. I’m not sure why Crin identified it as a modified SPV4040, which it is, and not a CB1200. Protocol Micro is (IMO) a nicer Hi-Max dongle, justifying the higher price (still very reasonable).

Side Note: TTGK Station in its latest version, is turning into an interesting test tool for these dongles, with information about the USB protocol, line control & mic detection, 3 signal generators that can be used independently or in combination (IMD test), white & pink noise generators, and a few other tests. In Chinese only (unfortunately).
TTGK also has an integrated Android-based version called “Audio-Z”. Could not find the list price anywhere.

 

[fws]

I’m not sure why Crin identified it as a modified SPV4040

I suspect because search results for CB1200AU will outright show the super cheap Hi-Max, while SPV4040 will likely show your post about $20 Snowsky Tiny, and worst case scenario my reply about discontinued JM11 for $8. So the association of the chip might make it to be perceived as pricier and sold under known brands.

 

[danadam]

Some "interesting" issue I noticed. I started with this seemingly not very demanding signal (in attachment):


It is 32-bit integer, 44.1 kHz sampling rate, 1 second long. It has 10 pulses, each pulse is 10 ms.

I resampled it with SoX to all regular sampling rates, up to 384 kHz (still in 32-bit integer). Then I played it through JCally JM20 and captured the output through RME Adi-2 Pro:

44k and multiples and 48k and multiples:
,

Here's a zoom to two pulses. 0.3 is about -30 dBFS, 0.005 is about -46 dBFS. The distortion is underlined with red:


The distortion disappears when before playing the file:

• 23-bit or more (i.e. 22, 21, etc) dither is added
• or high-pass, even at 0.1 Hz, is applied (e.g. using SoX and "highpass -1 0.1")

In the spectrograms above the left column is from files having 24-bit dither and the right column is from files having 23-bit dither.

 

[tccalvin]

Some "interesting" issue I noticed.

Cool.

I assume the OS wasn't resampling the higher sample rates?

Do you think this behavior will affect real audio content?

 

[danadam]

I assume the OS wasn't resampling the higher sample rates?

No, it wasn't. It was played through alsa's direct hardware access: aplay -D hw:...

Do you think this behavior will affect real audio content?

No, not really.

I checked the following test signals:


The problem happens only for impulse and pulse (so signals with high DC content) and deep silence afterwards. After such signal is played, it takes some time until the analog output of JM20 levels back to 0. And now comes a quirk of JM20. It seems to have some mechanism that shuts the output down (or something similar) when there is no sample larger than +/- 1 (in 24-bit domain) in the last 8192 samples. Note, larger than, not larger or equal. So 24-bit dither is not enough to keep the analog output on. And 8192 is a constant, independent from sampling rate, so for higher sampling rates the output is shut down sooner.

Here's the captured output (try to guess what was the sampling rate of the input file ), from top to bottom:

• impulse, 24-bit dither
• pulse, 24-bit dither
• impulse, 23-bit dither
• pulse, 23-bit dither

 

[tccalvin]

try to guess what was the sampling rate of the input file

was it 88.2k or 96k...?

 

[danadam]

was it 88.2k or 96k...?

96k. Eyeballing it, 8192 / (.112-.027) gives 96376.