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" <martin_green@snet.net> wrote in message 
news:1110405779.404112.212400@g14g2000cwa.googlegroups.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 Havm&#4294967295;ller. 

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$21754$edfadb0f@dread14.news.tele.dk>,
Leo Havm&#4294967295;ller <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 <wiml@hhhh.org>, Seattle, WA, USA. PGP keyID 27F772C1

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