TSWriter Continuity error every ~61s on 2nd HDD

[gclendon]

TV-Server Version: 1.0 RC2
MediaPortal Version: 1.0 RC2
MediaPortal Skin: default
Windows Version: XP SP3
CPU Type: Intel 2.8G
HDD: WD300G + SG500G + SG500G
Memory: 1.25G
Motherboard: Asus P4S133
Video Card: Sapphire HD3650
Video Card Driver: 8.530.0.0
Sound Card: Creative Audigy 4
Sound Card AC3: Coax out
Sound Card Driver: 5.12.4.1196
1. TV Card: DVICO Fusion
1. TV Card Type: DVB-T
1. TV Card Driver: 6.2.0.2
2. TV Card: Hauppage PVR150
2. TV Card Type: Analog
2. TV Card Driver: 2.0.35.23258
3. TV Card:
3. TV Card Type:
3. TV Card Driver:
4. TV Card:
4. TV Card Type:
4. TV Card Driver:
MPEG2 Video Codec: PDVD8
MPEG2 Audio Codec: PDVD8
h.264 Video Codec: PDVD8
Satelite/CableTV Provider: None
HTPC Case:
Cooling: CPU + chassis fans
Power Supply: Enermax EG301P-VE 300W
Remote:
TV: Panasonic
TV - HTPC Connection: Component

Hi!
I'm getting TSWriter continuity errors approximately every 61s when writing to all but one of my HDs. I understand that continuity errors are usually indicative of bad reception, but I don't get such a bad set of continuity errors if I have the recording destination set to the boot HD (the one holding Windows). If I set the destination recording drive to the boot HD, I still get continuity errors but they're more like a couple of times an hour or so.
Also, the regularity of the continuity errors is rather suspicious to me. The errors appear every 61s or 62s plus a bit.

All HDDs are set to ultra DMA mode 2.
I have tried moving the other two HDs around: trying one as a slave on the same controller as the boot HD; trying one as a master on the secondary controller; defragging the disk. I have set the tv server priority to high. None of these changes seem to have made a difference (if anything, it's worse!).

It's almost as if the non-boot HDs are too slow (which is not the case since I've written an entire TS stream to disk via "DVB Web Scheduler" without issue), or there's some magic PCI value that I need to set, or something... obvious.
I also have this vague idea that there was a recommended cluster size as well, but I can't seem to find that on the wiki. What should it be? (All HDs are currently 4k).
What am I doing wrong, and what else can I try?

I don't currently have direct access to the logs, but I'll post them when I can.

Many thanks,
Gene