Topping D10s + E1DA Cosmos ADC as a measuring system

[staticV3]

Again: everyone will see an iAP interface with a yellow exclamation mark in their Device Manager, regardless of which driver is used for the Cosmos ADC.

It is fully expected.

 

[Dumdum]

The one from the Cosmos ADC's product page: https://e1dashz.wixsite.com/index/cosmos-adc


That's normal. The iAP interface is some kind of workaround from Comtrue for Huawei devices and has nothing to do with the Cosmos ADC's operation.
It can safely be ignored.

Thanks for that, doing that install now, I had a newer version installed, but for trouble shooting I will install that version

 

[Dumdum]

Done and installled, just tested and still getting ‘bouncing’

I have also increased the buffers in rew to max to see if that helped any, it didn't

I am just running tests on the same Dsp with the focusrite 8i6 I have as that’s a hair cleaner than the 2i2 and you can just about see the harmonics down to 5th or 6th over the noise and get a SINAD of just over 90db which may be down a little due to noise in the interface I suspect

 

[Dumdum]

Again: everyone will see an iAP interface with a yellow exclamation mark in their Device Manager, regardless of which driver is used for the Cosmos ADC.

It is fully expected.

Sorry I was replying as you typed your initial post and hadn’t refreshed the page

 

[staticV3]

Done and installled, just tested and still getting ‘bouncing’

Bouncing is basically guaranteed using the Rectangular window and still likely using Flat top (depends on motherboard and USB cable quality IME).

All other windows should be stable.

 

[Dumdum]

Bouncing is basically guaranteed using the Rectangular window and still likely using Flat top (depends on motherboard and USB cable quality IME).

All other windows should be stable.

So not expected or good with Blackman Harris 4 or 7 then… any other ideas welcome

It does this with two completely different pcs

Lenovo yoga core i5

Hp which is a intel slow thing that’s 8yrs old

Regardless of pc I can’t take a stepped sine measurement without it jumping up before it gets to 250hz

It’s not so bad when you’re taking 1khz sine measurements for harmonics as you can let it run with 32 averages and hit save when it settles

 

[staticV3]

So not expected or good with Blackman Harris 4 or 7 then… any other ideas welcome

It does this with two completely different pcs

Lenovo yoga core i5

Hp which is a intel slow thing that’s 8yrs old

Regardless of pc I can’t take a stepped sine measurement without it jumping up before it gets to 250hz

It’s not so bad when you’re taking 1khz sine measurements for harmonics as you can let it run with 32 averages and hit save when it settles

Yeah I've had that happen once using a Ryzen Thinkpad, though annoyingly it just sorted itself out at some point.
At the time, my sweeps came out looking like this:


Could you try running LatencyMon on both of your systems?
That might tell us something about the potential culprit behind this instability.

 

[Dumdum]

Yeah I've had that happen once using a Ryzen Thinkpad, though annoyingly it just sorted itself out at some point.
At the time, my sweeps came out looking like this:
View attachment 346356

Could you try running LatencyMon on both of your systems?
That might tell us something about the potential culprit behind this instability.

 

[staticV3]

View attachment 346358

Googling wdf01000.sys brings up lots of forum posts describing latency and audio dropouts.

Some of these may have a solution that works for you.

Does running latencymon on your other PC give a similar result?

 

[Dumdum]

I don’t have the other pc handy at the minute, will carry on with experimentation tomorrow

Thank you for your time

 

[Dumdum]

I think the easiest course of action is to flatten the other laptop as it’s not the fastest and use that purely for measurements with a fresh install of windows 10, so that is my next plan

 

[Dumdum]

Reset the windows ten second pc with the intention of having it purely for testing… and removed all data bar the operating system and still bouncing… frustrating me this!

 

[Dumdum]

So I have got my issue a lot better, by chance I read of someone turning off some network drivers… so by chance I set the rta reading and turned off a bt device, and also the next device under that

I also have changed the power scheme to high performance, and changed the screen saver to 15 minutes on battery and 1hr on mains, this has also helped, I will try and introduce the correct Lenovo drivers for the two items I shut off and see what they do when reactivated with the Lenovo driver and not the windows driver

A positive night after all!

Got the noise floor measurement down to -123db also, and depending on the input sensitivity, had up to -111.6db for SINAD which isn’t far off the AP

 

[chuckt62]

Is the Cosmos Scaler an equivalent (+$$$) replacement for the Autoranger?

 

[staticV3]

Is the Cosmos Scaler an equivalent (+$$$) replacement for the Autoranger?

What Autoranger are you referring to?

 

[somebodyelse]

https://linearaudio.net/la-autoranger maybe?

 

[chuckt62]

Yes, that one.
It looks like the Scaler is more gain vs attenuation though.

 

[Dumdum]

No problem, I use usual resistor dividers, both SE and balanced configuration, at low impedance (not to add resistor noise), like 1k / 100R and appropriate switch setting on E1DA.

View attachment 211407

The SE divider + 2.7V EDA setting adds 13x attenuation, so the max. input is 35.1V. Simple math, you can make it as you like.

SE divider looks like this
View attachment 211408

The Dispre 2 - JFET preamp was measured directly, without a divider, in a setup as below:
View attachment 211409

So if I want to make a se divider like yours what values of resistors should be used to get roughly 5x the attenuation? So on the 10v setting I can use 50v input which is plenty for what I want to be measuring

I presume it has some reference in the equation to current and also voltage drop through the circuit

 

[lemmy_collins]

So if I want to make a se divider like yours what values of resistors should be used to get roughly 5x the attenuation? So on the 10v setting I can use 50v input which is plenty for what I want to be measuring

I presume it has some reference in the equation to current and also voltage drop through the circuit

R1/R2 should be 5 so 1k and 200 ohm should be OK but beware it will change by downing the setting because of the COSMOS input impedance, for 1.7Vrms the divider will be 6.6 because R2 will be in // with 640ohm input impedance.

 

[pavlo_y0l0s]

Hello everyone!

Just wanted to share my results using SMSL D6s + E1DA Cosmos ADC.

It took me a little bit of time to figure out how to get both of them selected in REW preferences. Just installed FlexASIO and selected that as ASIO device.

Measured loopback by plugging XLR cable into both devices' right channels.

REW generator was setup to 1kHz 0dBu level (774Vrms). Volume on D6s DAC was set to max.

I am getting 0.00017% THD+N and 0.000046% THD.