Geshelli Archel2 Headphone AMP Review: another champ?

[Jave]

The more I look at the review and then the THD vs frequency the more the first measurements seem unlikely. There is a huge difference in THD between the channels and the overall THD+N goes up 10 times the original 1KHz measurements.

This smells bad

@amirm - if you could post some of the measurements in 800Hz or 1.2KHz (or some other random frequency), it will be much appreciated. Thanks!

 

[Schackmannen]

The more I look at the review and then the THD vs frequency the more the first measurements seem unlikely. There is a huge difference in THD between the channels and the overall THD+N goes up 10 times the original 1KHz measurements.

This smells bad

@amirm - if you could post some of the measurements in 800Hz or 1.2KHz (or some other random frequency), it will be much appreciated. Thanks!

The THD+N vs frequency test uses a much wider bandwidth vs the dashboard so it will always be worse than the SINAD in the dashboard.

 

[Jave]

Exactly why I asked for the same measurements in different frequency, to see whether the noise is ultrasonic, and therefore irrelevant, or whether it is a real problem.

Even the same THD+N vs frequency with a smaller bandwidth will be very helpful

 

[Jave]

Why has the thread gone silent all of a sudden?

 

[solderdude]

Maybe all is said and done until it actually becomes available and then user reports will be coming in.

800Hz and 1.2kHz, measured the same way as the 1kHz tone, will produce the exact same results as the 1kHz plot.
It is too close to 1kHz to make a difference.
And no... you can't build an amplifier that gives near perfect measurements at 1kHz and doesn't at 800Hz and 1.2kHz without using an extremely sharp notch filter at the input of the amp and not be captured in the FR plot.
The 'lesser' performance plots are already shown which show that noise above 20kHz skews the results and shows inaudible rise of noise at the lowest part of the FR.

 

[Jave]

So still, seeing a THD+N vs frequency for a narrower band will probably show useful information, hopefully not masked by high-frequency THD\N.

 

[solderdude]

It's below any audible thresholds... why should one worry. It has been explained where the higher THD numbers come from.

The plot below tells the whole story. All distortion productsare shown including 800Hz and 1200Hz (or very close to it) and I put the audibility threshold in it. Everything below the 0dB limit is no of any consequence. It can be measured but not heard.

 

[maty]

Maybe you can pull out the cheap-ass Chinese pot and put in an Alps of your choice with a $1 adapter PCB.

https://www.ebay.com.au/itm/Adapter...m4b4b4848f9:g:tCUAAOSwJ8pbaX21&frcectupt=true

View attachment 36804

You may be able to squeeze it in, or not.

I trust Aleksandar's products more. From Bulgary, EU.

https://www.atlhifi.com/shop/bare-pcb/alps-blue-velvet-pcb/ € 2

[ PCB is with dimensions 28×32.5 mm (1.1×1.28 in.).

PCB thickness: FR4 1.55 mm. Copper thickness: 35um (1oz). Surface finish – Nickel Gold (ENIG).

Now it is TWO-layer design with solder mask coating on top and bottom layers for best durability. Silk screen is placed on the top of the PCB. The whole top copper layer is used as a ground plane for better shielding.

PCB has abilities for mounting 1 pcs. ALPS RK27112 (Blue velvet) and 2 pcs. 4-pins Molex Headers & Wire Housings (Mouser part # 538-171856-0004) for INPUT and OUTPUT connectors. ]

 

[Jave]

It's below any audible thresholds... why should one worry. It has been explained where the higher THD numbers come from.

The plot below tells the whole story. All distortion productsare shown including 800Hz and 1200Hz (or very close to it) and I put the audibility threshold in it. Everything below the 0dB limit is no of any consequence. It can be measured but not heard.

View attachment 37090

I'd still like to know what are the harmonics which pop up at low frequency

 

[solderdude]

Maybe the DCDC converter, maybe a less dielectric quality than X7R ceramic coupling caps are used... one would have to own an AP and some parts to experiment and find out. In any case... it is not of any importance as it is below audible thresholds and we are talking about a relatively cheap amp in a similar price class as the Atom.

 

[bequietjk]

Can't wait to get mine

In the meantime, my Archel Pro is getting adjusted from 6x gain to 2x gain. Curious to see how it plays with planars (t50rp)

 

[bigjacko]

Can't wait to get mine

In the meantime, my Archel Pro is getting adjusted from 6x gain to 2x gain. Curious to see how it plays with planars (t50rp)

Do you think archel 2 can power t50 rp fine? How can I tell if it powers fine?

 

[Schackmannen]

Do you think archel 2 can power t50 rp fine? How can I tell if it powers fine?

If it gets loud enough for you without clipping.

 

[bequietjk]

Do you think archel 2 can power t50 rp fine? How can I tell if it powers fine?

I have heard Archel1 and it gets plenty loud. We will have to wait and see for the Archel2 but I suspect it should do just fine

 

[bigjacko]

I have heard Archel1 and it gets plenty loud. We will have to wait and see for the Archel2 but I suspect it should do just fine

Do they have the same power output?

 

[bequietjk]

On Geshelli's website it states the Archel1 having 1W@16ohms whereas the Archel2 offers 500mW@16ohms

 

[pglee]

I don't need this, but at this price, I couldn't resist. Much thanks to this site, I think we're in the golden age for headphone audio.

 

[alecm]

new balanced amp from geshelli coming soon. more powerful ...

 

[Tks]

new balanced amp from geshelli coming soon. more powerful ...

I hope they decide against those release tabs. Those things are more trouble than they're worth. But then again unlike most companies as of late, these are Neutrik, and should be good in that case.

 

[DualTriode]

Hello All,

I am new here.

This Geshelli Archel2 Headphone is looking good. I just placed a pre-order for one.

PG&E has been turning off our power here in northern California lately. I am looking to assemble a battery powered listening system to use in the dark. There should be no 60 cycle hum leaking in.

I do notice some power frequency and related harmonics in the posted dashboard FFT. Minus the power supply noise the SNR should really pop up there.

Amirm I am surprised that you did not use the highest performance mode of your analyzer.

Looking forward to posting thoughts and plots.

Thanks DT