Msg for Rudolf Usselmann

S

SneakerNet

Guest
Hi Rudolf

This is with regards to your usb 1.1 function core from opencores.org.

Your USB clock needs a 48MHz clock speed. I have a Nios Development Board
with a 50MHz oscillator. I'm using a PLL to drop the speed to 48.076923MHz
(ratio of 25/26). Is that OK or does it have to be ditto 48MHz cuz that's
the best that I can get.

Regards
 
SneakerNet wrote:

Hi Rudolf

This is with regards to your usb 1.1 function core from opencores.org.

Your USB clock needs a 48MHz clock speed. I have a Nios Development Board
with a 50MHz oscillator. I'm using a PLL to drop the speed to 48.076923MHz
(ratio of 25/26). Is that OK or does it have to be ditto 48MHz cuz that's
the best that I can get.

Regards
Click to expand...

I don't know from the top of my head, you need to check
the USB spec. Look at the max error rate they allow.

Please post follow up messages to the usb forum at OpenCores.org.

Thanks,
rudi
========================================================
ASICS.ws ::: Solutions for your ASIC/FPGA needs :::
...............::: FPGAs * Full Custom ICs * IP Cores :::
FREE IP Cores -> http://www.asics.ws/ <- FREE EDA Tools
 
"SneakerNet" <nospam@nospam.org> wrote in message
news:ILMac.1374$d%6.39465@news.xtra.co.nz...
Hi Rudolf

This is with regards to your usb 1.1 function core from opencores.org.

Your USB clock needs a 48MHz clock speed. I have a Nios Development Board
with a 50MHz oscillator. I'm using a PLL to drop the speed to 48.076923MHz
(ratio of 25/26). Is that OK or does it have to be ditto 48MHz cuz that's
the best that I can get.
Click to expand...
it will receive 100% ok also with 50MHz clock
I once tested accidently, not sure if the hosts will accept 50MHz transmit
48.08 will defenetly work both ways, not matter if it is withing spec range
or not

Antti Lukats
 
Thanks Rudolf and Antti

"SneakerNet" <nospam@nospam.org> wrote in message
news:ILMac.1374$d%6.39465@news.xtra.co.nz...
Hi Rudolf

This is with regards to your usb 1.1 function core from opencores.org.

Your USB clock needs a 48MHz clock speed. I have a Nios Development Board
with a 50MHz oscillator. I'm using a PLL to drop the speed to 48.076923MHz
(ratio of 25/26). Is that OK or does it have to be ditto 48MHz cuz that's
the best that I can get.

Regards
Click to expand...
 
Antti Lukats <antti@case2000.com> wrote:

Hi Antti!
Your USB clock needs a 48MHz clock speed. I have a Nios Development Board
with a 50MHz oscillator. I'm using a PLL to drop the speed to 48.076923MHz
(ratio of 25/26). Is that OK or does it have to be ditto 48MHz cuz that's
the best that I can get.

it will receive 100% ok also with 50MHz clock
I once tested accidently, not sure if the hosts will accept 50MHz transmit
48.08 will defenetly work both ways, not matter if it is withing spec range
or not
Click to expand...
At work we have also worked on USB devices which are connected to a PC.
In this case the tolerance of ceramic resonator can be too much. I am
not sure about the exact figures, if I am not wrong the spec is about
+/-100ppm or slightly more. In some application notes you can find also
statements like "should not work with ceramic resonators but no problems
in practice". So using a 50MHz clock does not sound too much promising
for me.

Best regards,

Christoph
>
 
"Christoph Brinkhaus" <c.brinkhaus@t-online.de> wrote in message
news:4enu4c.hl.ln@lola.de...
Antti Lukats <antti@case2000.com> wrote:

Hi Antti!

Your USB clock needs a 48MHz clock speed. I have a Nios Development
Board
with a 50MHz oscillator. I'm using a PLL to drop the speed to
48.076923MHz
(ratio of 25/26). Is that OK or does it have to be ditto 48MHz cuz
that's
the best that I can get.

it will receive 100% ok also with 50MHz clock
I once tested accidently, not sure if the hosts will accept 50MHz
transmit
48.08 will defenetly work both ways, not matter if it is withing spec
range
or not

At work we have also worked on USB devices which are connected to a PC.
In this case the tolerance of ceramic resonator can be too much. I am
not sure about the exact figures, if I am not wrong the spec is about
+/-100ppm or slightly more. In some application notes you can find also
statements like "should not work with ceramic resonators but no problems
in practice". So using a 50MHz clock does not sound too much promising
for me.
Click to expand...
Sure,
I just said that the opencores usb1_phy DPLL tolerates 50MHz on the receive
side!
as I by accident did figure out.

nothing more. for any serious projec the specs should be followed of course!

antti
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top