Squeezebox Classic - Radio Clock Swap

[chrism]

Anyone swapped out the internet radio xtal with a low jitter clock? I have a 24 Mhz clock made for me but struggling with installing it.

I can see where the 5v supply goes but not sure about the grd wire and clock signal wire to the HCU04AG.

Can anyone help?

Regards

Chris

[leo]

24mhz? you sure :scratch: I think its more around 12mhz for the radio

[chrism]

Hi Leo,

Looks like you are right. One clock appears to be 11.28 (Y1) and the other 12.28Mhz (Y2). I don't think that the 12.28 one can be changed on it's own as I am proposing. Looks like both clocks need swapping to gain improvements to the radio.

Regards

[ar-t]

Supposedly.........

The 12.288 MHz controls the radio function. However, I have ripped it out of several units, and the owners claim radio function still works.

You may want to more thoroughly investigate what the crystals do, before plunking down money on a special clock/crystal.

Anyway.............why do you think changing that crystal will help things? In theory, installing a low phase noise unit will improve things. However, if all you do is stick it into the chip that functions as the clock, you are not going to gain a damn thing. It takes a lot more than that, to make a difference.

[chrism]

Ar-t,

Thanks for your help but I must say the message has flown straight over my head and I can't see what help it is to me.

I am proposing to try a C1 type clock and I understand that the Y2 crystal affects some radio frequencies.

Regards

[leo]

Doesn't the 11.2896MHz XO deal with 44.1k material and the 12.288MHz 48k?

With the clock I tried it runs the CPLD direct , the HCU04 was left in the SPDIF out.
TBH I'm not sure if a clock upgrade would help much with the radio's quality, I thought it sounded a lot like MP3 anyway

[ar-t]

Ar-t,

Thanks for your help but I must say the message has flown straight over my head and I can't see what help it is to me.

Yes, I know that. It was intentional.

Changing clocks and/or crystals, with the hope of making things better, is a lot more involved, than you might think.

I'll try to make this reply, a bit easier to grasp, but just as discouraging.

There is one chip, in that unit, that does the following:

11.2896 MHz clock
12.228 MHz clock
SPDIF driver.

First, any clock that runs through that chip is going to be corrupted, by the SPDIF signal. And it will be corrupted in a nasty manner, that can not be cleaned up, by any method. So, you are wasting your money, if you spend money on a good clock/crystal, and shove it back into that chip.

You would also need to separate the 2 clocks. Two of them, operating at the same time, is not good for sound.

Assuming you can get around those 2 problems, you then have to come up with a really clean supply, for the clock. The internal supply is far too noisy, and you will not have much success, in cleaning them up, to make them acceptable.

Now, before someone accuses me of slagging this product, don't even think of it. For the price, they pack a lot of performance, into a small package, at a small price. To remedy those issues............well, you will notice they have a much more expensive product.


I am proposing to try a C1 type clock and I understand that the Y2 crystal affects some radio frequencies.

I have to admit I have no knowledge of a "C1 type clock", but assuming it is a good unit, it will not yield better performance, without a lot of redesign.

Does that make any more sense?

Pat

[ar-t]

Doesn't the 11.2896MHz XO deal with 44.1k material and the 12.288MHz 48k?

Yep, that is way it is supposed to work. However, I have modified many units, and the owners claim they still get radio. Now, since the 48 kHz clock is gone, one can only assume that big chip somehow figures out how to resample radio signals.


With the clock I tried it runs the CPLD direct , the HCU04 was left in the SPDIF out.

That will work. I hope you disabled the parts of the -U04, that did that clock function.

Pat