Philips PDIUSBD12 Cannot Poll the interrupt Register

Hi,
I am using the Philips PDIUSBD12.
I am polling the device since I cannot use the interrupt pin.
The part seems to work most of the time though it occassionally gives
bad IN packets during enumeration.
I also can only poll the interrupt register once after enumeration to
see if I am getting an OUT packet. If I poll the interrupt register
after enumeration then I start sending bad or invalid IN packets, does
anyone have any idea what might cause the bad packets after
enumeration?

Thanks,
Martin Green

"MartinGreen" <> wrote in message
news: oups.com...
> Hi,
> I am using the Philips PDIUSBD12.
> I am polling the device since I cannot use the interrupt pin.
> The part seems to work most of the time though it occassionally gives
> bad IN packets during enumeration.
> I also can only poll the interrupt register once after enumeration to
> see if I am getting an OUT packet. If I poll the interrupt register
> after enumeration then I start sending bad or invalid IN packets, does
> anyone have any idea what might cause the bad packets after
> enumeration?

This chip is very buggy and a royal PITA to work with. I have written
drivers for 4 different USB slave chips and this one is just horrible.
But first and formost: Do you have the address/data bus timing right? It
needs some very long delays (500-600 ns). See table 17 and figure 18 in the
datasheet.

Leo Havmøller.

Hi,
Thanks for writing back.
I did as you asked and checked the timing, I am using an embedded 8051,
the fastest I can do anything is ~500uS and I looked at the long
delays, I am well within spec.
The wierd thing is that I can do all the enumeration (although I get
bad packets), but if I poll the interrupt register it gets messed up.

I appreciate your response, maybe you can shed some more light on this
problem.

Thanks,
Martin

In article <422fde9b$0$21754$> ,
Leo Havmøller <leh@-nospam-rtx.dk> wrote:
>This chip is very buggy and a royal PITA to work with. I have written
>drivers for 4 different USB slave chips and this one is just horrible.
>But first and formost: Do you have the address/data bus timing right? It
>needs some very long delays (500-600 ns). See table 17 and figure 18 in the
>datasheet.

Interesting to hear. I was just looking at the PDIUSBD11 or -12 for
something. Maybe I'd better look at some more alternatives. Do you
have better experiences with some other USB slave chips of this
sort?

--
Wim Lewis <>, Seattle, WA, USA. PGP keyID 27F772C1

Hi,
unfortuneatley this is the only chip I have tried to act as an external
usb device.
Martin