Metronome guidance. 
A couple of notes,
Notice that unless you disable the on-board clock, SCK(the master clock) is ALWAYS an output on the Metronome for both PCM IN and PCM OUT. It is there to allow for use of both master and slave modes on input and output. You can disable the XO on board to allow the master clock to be driven externally (pull pin one of the XO to GND), but this is not a standard configuration.
You will almost never want to connect the SCK from your PCM source to SCK on the PCM input side!!! You will want to use it on the output side of course.
Here is an image of a correctly configured metronome configured for 24bit I2S 192khz.
Please Please Please notice that some of the pin headers are not jumped intentionally. Just jump the one I have jumped. The other pin headers are there to allow people who wish to use micro-controllers to configure and monitor the device. Only use them if you have read the data sheet and know what you are doing.
Cheers!
Russ
A couple of notes,
Notice that unless you disable the on-board clock, SCK(the master clock) is ALWAYS an output on the Metronome for both PCM IN and PCM OUT. It is there to allow for use of both master and slave modes on input and output. You can disable the XO on board to allow the master clock to be driven externally (pull pin one of the XO to GND), but this is not a standard configuration.
You will almost never want to connect the SCK from your PCM source to SCK on the PCM input side!!!
You will want to use it on the output side of course.Here is an image of a correctly configured metronome configured for 24bit I2S 192khz.
Please Please Please notice that some of the pin headers are not jumped intentionally. Just jump the one I have jumped. The other pin headers are there to allow people who wish to use micro-controllers to configure and monitor the device. Only use them if you have read the data sheet and know what you are doing.
Cheers!
Russ
Attachments
ASRC Impressions and questions
Having fun with my ASRC. Looking at the data sheets, one can run this as just a reclocker, w/o oversampling, correct? BTW, is the Metronome ASRC designed to be fed 7 or 5v?
What I tried-
MODE 0, 1 and 2 to= lo-hi-lo (to 512 fs) -does that reclock to 48 khz?
IFMT 0, 1 and 2= stays at hi, lo, lo (for i2s)
OFMT 0 and 1= stays at hi, lo (for i2s)
OWL 0 and 1= hi, hi for 16 bit
Above seems to work- I get sound!
I've attached some data sheets pulled from the src4192 evaluation board guide. Maybe it can help some other newbies like myself!
First- I am experiencing slight, soft 'clicks' or pauses occasionally. At 128 fs, they occur every minute or two; and very soft. At the new settings above 512 fs they occur more frequently. Could this be buffer under-run on the SRC4192 or a re-sync with incoming data? The stream via usb from my computer is not slaved to the ASRC, correct?
ASRC first impressions- I have been listening to the OPUS sans ASRC for about a week now. System- Mac usb>opus>ballsie>Asusa 845 SE mono-blocks>Yamaha NS1000's. Using itunes to control volume. Not the best, but fine for now. Consistently surprised how nice this little dac sounds.
So I added the ASRC yesterday. Much more focused and 3d- noticing more wide sound field. But the highs are too 'tizzy' and 'splashy' for me. Mid range less full. Subjective, I know. I will need to settle in with it more.
Having fun with my ASRC. Looking at the data sheets, one can run this as just a reclocker, w/o oversampling, correct? BTW, is the Metronome ASRC designed to be fed 7 or 5v?
What I tried-
MODE 0, 1 and 2 to= lo-hi-lo (to 512 fs) -does that reclock to 48 khz?
IFMT 0, 1 and 2= stays at hi, lo, lo (for i2s)
OFMT 0 and 1= stays at hi, lo (for i2s)
OWL 0 and 1= hi, hi for 16 bit
Above seems to work- I get sound!
I've attached some data sheets pulled from the src4192 evaluation board guide. Maybe it can help some other newbies like myself!
First- I am experiencing slight, soft 'clicks' or pauses occasionally. At 128 fs, they occur every minute or two; and very soft. At the new settings above 512 fs they occur more frequently. Could this be buffer under-run on the SRC4192 or a re-sync with incoming data? The stream via usb from my computer is not slaved to the ASRC, correct?
ASRC first impressions- I have been listening to the OPUS sans ASRC for about a week now. System- Mac usb>opus>ballsie>Asusa 845 SE mono-blocks>Yamaha NS1000's. Using itunes to control volume. Not the best, but fine for now. Consistently surprised how nice this little dac sounds.
So I added the ASRC yesterday. Much more focused and 3d- noticing more wide sound field. But the highs are too 'tizzy' and 'splashy' for me. Mid range less full. Subjective, I know. I will need to settle in with it more.
The metronome has an on-board LDO regulator at 3.3V so 5V or 7V is supply just fine.
Yes, 512fs = 48khz (with the XO on the board).
No you should not be getting any over/under run. I have never experienced the sound you are describing. Especially no "Tizzy" or "Splashy" ness.
You should probably be using 24bit rather than 16 for output word length.
I am glad you like it.
Cheers!
Russ
Yes, 512fs = 48khz (with the XO on the board).
No you should not be getting any over/under run. I have never experienced the sound you are describing. Especially no "Tizzy" or "Splashy" ness.
You should probably be using 24bit rather than 16 for output word length.
I am glad you like it.
Cheers!
Russ
kstlfido said:Thanks, Russ! You bet I like it!
My apologies- the sounds I was hearing seem to be hard drive related.
ASRC sounds very good indeed!!!
Excellent.
Since the DAC is 24bit I would go ahead set the output word length to 24bit.
Also, the positive effects of the ASRC will be more pronounced the farther removed from the input frequency. 96khz (256fs) works well, and I think (personal subjective opinion) 192khz (128fs) gives a very nice effect. I would say the ASRC seems to "polish" things up a bit, not adding anything, just making what is there more appealing.
This could be completely in my head.Cheers!
Russ
Russ White said:I would say the ASRC seems to "polish" things up a bit, not adding anything, just making what is there more appealing.
The metronome arrived yesterday in perfect condition. (as always with your modules! thanks!)
My son came over to listen to some of the recodings I made of him playing the drums.
(recording was 16/48 DAT, converted to 16/44.1 and burned on CD)
We started with my original setup: cd>rec>dac
He found the sound very accurate but commented on the bass not being tight enough and the highs not bright enough.
(I'm listening to this setup for two months now and understand his comment, but I still like the sound)
So I hooked up the metronome: CD>rec>ASRC>dac
Configured it as master and 24/128fs output.
His comment was very harsh: "this is not my drumkit"
We listened to other tracks and yes, I found it true, sound is improved, tighter bass, brighter high, but far less accurate and even a bit synthetic. (could be compared to a, mp3 encoding)
Soundstage had become wider but the center was a bit jumpy...
Piano's get turned into high quality synthesisers
We decided to stick with 128fs and try different output sample rates. This helped... A LOT.
We did blind tests but every time I hit 16 bit, he noticed it immediately. (one cannot fiddle with jumpers and listen at the same time) I took the time to listen too and I have to say, 16bit was the most accurate setting.
A bit strange, since every output word length is dithered from the internal 28-bits.
Finally, we compared with the original setup, took some time because I didn't want to use the bypass-function.
The configuration with the metronome at 16/128fs easily beat the original configuration at every criterium. (soundstage, bass an highs)
When we have more time, we'll check out the other output frequency options. Maybe we come to different conclusions.
Our ears were tired enough after a day like this
In the meantime, I'm enjoying the music, and that was an important part of the reason to build this thing
btw, a little question...
The receiver board eats 56mA.
DAC VD 18,6mA standby, 20,3mA playing
DAC VA 11,6mA
All these values are normal?
Is there a simple way of disabling all the LED's on the receiver?
The transformer is feeding 2x16V into the LCPS and the digital side is a tad too hot to my liking, output set to 7.5V.
LCPS itself is perfectly fine.
Hello OneyedK,
Yes the current draws seem within what I would expect.
I would not really be worried about the heat sink we made them pretty beefy. Unless its too hot to comfortably touch I would not worry in the least. It is ok for it to run very warm.
How do you have your other components configured? this could be why you found a difference.
For reference, I have my receiver and my DAC configured for 24bit I2S. I have found that the 24bit I2S output setting of the metronome seems to be the most effective for me. I have been switching between 256fs and 128fs (96khz and 192khz) and find both to be very good. I can't really pick a clear favorite yet. More time maybe.
The only way to disable the LEDs would be to disconnect them, I don't think you really need to. The LCDPS should be more than able to handle your setup.
Cheers!
Russ
Yes the current draws seem within what I would expect.
I would not really be worried about the heat sink we made them pretty beefy. Unless its too hot to comfortably touch I would not worry in the least. It is ok for it to run very warm.
How do you have your other components configured? this could be why you found a difference.
For reference, I have my receiver and my DAC configured for 24bit I2S. I have found that the 24bit I2S output setting of the metronome seems to be the most effective for me. I have been switching between 256fs and 128fs (96khz and 192khz) and find both to be very good. I can't really pick a clear favorite yet. More time maybe.
The only way to disable the LEDs would be to disconnect them, I don't think you really need to. The LCDPS should be more than able to handle your setup.
Cheers!
Russ
Some things to keep in mind...
For the WM8740(Opus) operating at 128fs makes the device switch from its usual 128X oversampling to 64X oversample and it bypasses the first stage of digital filtering.
This may or may not be noticeable.
Notice in the datasheet (Page 18) that the filter response in 192khz mode is different.
Cheers!
Russ
For the WM8740(Opus) operating at 128fs makes the device switch from its usual 128X oversampling to 64X oversample and it bypasses the first stage of digital filtering.
This may or may not be noticeable.
Notice in the datasheet (Page 18) that the filter response in 192khz mode is different.
Cheers!
Russ
I've recently come across a head unit for my car that has a digital out and I thought it might be worth a try to hook it up to my opus to see how it sounds. My question is what would be the best way to set it up for DC power? My assumption is that the LCPS won't work with the 12v DC that's provided by the car since it's designed for AC voltage from the transformer in normal usage? Thanks for any help!
neb001 said:
Actually it will work just fine. You just won't need a transformer. Just connect GND to one of the AC input, and 12V (its actually probably higher like 14.5V) to the other
The diodes will take care of it. If your voltage is < 16V you actually would probably not need LCDPS at all for the DAC. But I have not tested that (I have not had any reason to).
Cheers!
Russ
Russ White said:
Actually it will work just fine. You just won't need a transformer. Just connect GND to one of the AC input, and 12V (its actually probably higher like 14.5V) to the other
If your voltage is < 16V you actually would probably not need LCDPS at all for the DAC. But I have not tested that (I have not had any reason to).
Cheers!
Russ
That's what I was hoping for, but opted to ask rather than find out the hard way it doesn't work
Thanks Russ!
Russ White said:
Fiddled a bit more with the settings today...
(spent some time reading those darn datasheets too)
So First the receiver...
AIF_MS : 1 : MCLK, LRCLK and BCLK have to be generated from the S/PDIF input
TXSRC : 1 : sets the source of the S/PDIF transmitter to the reconstructed S/PDIF input data
(not important now, but I want to do some measurements later)
AIFCONF1, AIFCONF0 : 0,0 : 16 bit (same as source, CD-player)
The other possible setting : 0,1 : is pretty useless if I interpreted the note in the datasheet correctly:
Datasheet WM8804 page 51 note 2:
"In 24 bit I²S mode, any data width of 24 bits or less is supported provided that LRCLK is high for a minimum of 24 BCLK cycles
and low for a minimum of 24 BCLK cycles (48 BCLK cycles).
If exactly 32 BCLK cycles occur in one LRCLK (16 high, 16 low) the chip will auto detect and operate in 16 bit data word length mode.
So, 16/44.1 I²S will be the input for the metronome...
Metronome jumpers:
(have a look at the picture please, I might have overlooked something)
An externally hosted image should be here but it was not working when we last tested it.
Output port should be master, and I'd like it to be 256fs:
Mode2 : 0
Mode1 : 1
Mode0 : 1
Input port data format is 16-bit I²S (this seems to be okay for the 24-bit I²S input setting
according to the input port operating description on page 18 of the datasheet)
IFMT2 : 0
IFMT1 : 0
IFMT0 : 1
The output data port format should be I²S:
OFMT1 : 0
OFMT0 : 1
Since I lowered the output samplerate to 256fs, I thought I'd give the 24-bit output data word length another go:
OWL1 : 0
OWL0 : 0
For the rest:
Mute : 0 : I have no interest in a mute of the output ;-)
RST : 1 : No need for a reset either
Bypass : 0 : LOL, I really don't want to buy a module to bypass it afterwards
LGRP : 0 : -blush- don't understand that one
RDY : open : an output, I don't want to jumper that one
Now for the DAC-module itself, since it's in hardware-mode, we don't have many options (not even the slow-rolloff switch!)
I²S : 1 : Yes, I do want I²S input
DM1 and DM0 : 0 and 0 : all other options enable some form of de-emphasis (I thought I'd be able to implement slow rolloff, but that's definately NOT true)
MUTEB : 1 : Mute off
MODE : 0 : Hardware mode selected (better not change this one)
IWO : Input word length : bit tricky this one, if you set it to 0, only I²S 16-bit signals will be accepted,
If you set it to 1, according to the datasheet, 20- and 24-bit I²S signals will be accepted.
(but I tried feeding it with 16-bit I²S signals and it worked flawlessly)
Didn't connect RSTB and ZERO jumpers.
DIFFHW : 0 : I only have one DAC
M8X : 0 : -blush- another one I don't understand, setting it to 1 simply didn't work...
So, did I miss any other options I'd want to try???
This setting sounds great imho, I think it (24-bit/256fs) sounds a tad more subtle than the 16-bit/128fs I liked yesterday.
Will listen to it for a while, then give the 24-bit/128fs-setting another go.
OneyedK said:
Leave RESET,RDY,MUTE open on the metronome, also make absolutely sure you do not use the SCK out from the receiver into the metronome.
Setting those will screw up the power on reset and automute functions of the metronome.
Look at the pic I posted here:
http://www.diyaudio.com/forums/showthread.php?postid=1363864#post1363864
The most flexible way to setup the DAC/REC/Metronome is 24bit I2S, if you have a 16bit source the low bits are simply tossed out, or the device falls back to 16bit.
Cheers!
Russ
Dougie085 said:Ohhhhh....so it is something I may want
Do you think it would help USB use a lot as far as jitter and what not?
Yes, it does clean up the USB input very very nicely.
While the PCM output of the USB module on its its own is certainly not bad, it is better with the metronome. The cool thing is that the metronome can be operated off the VBUS. No additional power supply required.
Cheers!
Russ
Russ White said:
Okay, since finding time to actually build the LCDPS is what is holding up all my opus experiments, this intrigues me. I had thought that the onboard regs would be very sensitive to dissipation, and latched on to your suggestion of setting the LCDPS to 7.5V as an indication of that. If they can handle 12V then I can simply throw one of my existing SLA batteries into the mix and get started.
I may do a gut-check and try it, since I don't have a good soldering setup at the moment. Do I get put at the front of the line for re-orders if I blow something up?