Well, it seems I'm taking "Everything Engineering" to the very limits of=20
on-topicness once more :)

I've put together a CDCE925-based oscillator on a Roth Elektronik=20
TSSOP16 prototyping board. This consists of the chip, a 13.876MHz=20
crystal, and a bunch of wires. The programmer I'm using is a homebrew=20
version of the TI CDCE925 Performance Evaluation Board programmer (page=20
  24). I2C pullups are 2x10k to 1.8V (though I've tried 3.3V as well).=20
I'm using TI's software (TI ClockPro) to program the chip... or at least=20
I'm trying to.

The  problem I'm having is that while ClockPro can see the programmer=20
pod --  there are start pulses and address bytes going over the I2C bus=20
-- the  chip isn't ACKing the address bytes. This means ClockPro doesn't=20
list it  as an available chip -- or rather it tells me that I should=20
pull VDDOUT  to ground. I have already done this (in fact, I've tried=20
grounding and  connecting to +3.3V) and the chip is still ignoring=20
everything I send  it. If I short SDA to ground, then ClockPro detects=20
"something", and promptly has kittens when it tries to figure out what=20
I've done...

I've checked the signalling on my scope (200MHz Tek TDS2024B  DSO) and=20
it looks fine -- nice square edges, minimal=20
ringing/overshoot/undershoot. I measured the bus clock frequency at=20
about 120kHz.

All grounds and power lines are stable, and the  power looks good (1.8V=20
on VDD, provided by a NatSemi LP5951MF-1.8, 3.3V  provided by my bench PSU)=
..

I've tried swapping SDA and SCL, and even tried a new chip. Same issue.

It looks to me like the chip isn't pulling SDA low, or isn't recognising=20
the start sequence / address byte...

Does anyone have any idea what might be going on here?

Thanks,
--=20
Phil.
piclist@philpem.me.uk
http://www.philpem.me.uk/
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.