End of weekend update

[pHaestus]

Ok this weekend I did manage to get a few things done, but with puzzling results. First of all, I set up a test system with an MSI NForce motherboard. I got the Maxim MAX6655EVSys running on it, but the remote sensor for CPU reported 70C. This was odd, because it only ranged from 70C to 73C under load, while the in socket thermistor changed from 35 to 51 or so. My best guess is that MSI has some circuitry on the diode pins that is affecting the reading. The remote sensor worked properly when connected to a 3904 transistor. MBM and Speedfan picked up a third sensor onboard that also was at 69-70C. I DON'T think the board properly reads from diode, as this reading was unchanged when I removed the maxim hardware.

So I decided to set up the Epox 8k7a instead. Apparently it is a dead mobo now (I can find nothing wrong with it but just get FF at post). Tried changing CPU, RAM, and PSU. I had a PCI vid card and a SCSI hdd in it before; it may be that the agp port has some issues and a pci card will get it working. I am not holding my breath.

One thing I DID get done was a little work on a Thermaltake case. This should be familiar to most as a Chieftech/Antec case.

Here are a couple of pictures:

http://www.procooling.com/~phaestus/bipromount.jpg
http://www.procooling.com/~phaestus/bipromount2.jpg

and



I still have to redo a few things (shroud for radiator, gonna use difft tubing) and I have to wire the pump, but it is moving right along. This will probably be used in the living room for watching divx, listening to mp3s, and interfacing with my wife's digital piano.

[pHaestus]

Found the problem: A surface mount component got knocked off when I used vise grips to unstick some Gooped on swiftech standoffs. I was sure this was a big deal, but after desoldering another "LA" labeled component and measuring 0 ohms resistance I just jumpered the two pads. The Epox seems to work just fine again now. Huzzah!

I will maybe get a chance to mess around some tomorrow with the testing. A quick first article and then some cross testing with one of Bill's blocks.

Oh and I redid the res on my new system:

http://www.procooling.com/~phaestus/ttcase2.jpg

In fact, it will end up even difft from that.

[BillA]

Originally posted by pHaestus
. . . .
Oh and I redid the res on my new system:

http://www.procooling.com/~phaestus/ttcase2.jpg

In fact, it will end up even difft from that.

hope so, max out that pump inlet
is that a pressure-tight top on the res ?
what is it ? (how does it affix, threads and an o-ring ??)

and feed the res at the upper fluid level, or just slightly below

give me some warning on the wb you want as I'll have to re-run it;
-> am in the midst of HUGE changes in how I inspect flatness, and achieve it

[pHaestus]

I still have to tap a hole for the inlet fitting on the side, but this should be a good bit better:



As for which block, I don't think we have many that overlap. Your choice of a block you no longer need?

[BillA]

nice pump inlet
and the top ? how does it seal ?

re the wb, all we need is a comparison - right ?
the LiquidCC Surge ?

[pHaestus]

I cut threads off of a bushing and "gooped" them to the clear acrylic. There is a bead of dried goop all around the top to hopefully make a seal. If I am lucky, I might find an o ring the same size.

LiquidCC Surge is fine by me. Think you'll get it retested in the next couple of weeks? I will be in northern Louisiana over Christmas, so it would be a good bit cheaper to send it to there.

[pHaestus]

Success! I got the diode reader all soldered in and wired up on the 8K7A. I am about to grab a beer and then mount a heatsink for trial 1. It only took 5 days

Have I posted any pics of maxim's pcb?

http://www.procooling.com/~phaestus/...evsyswired.jpg

(big picture, sorry)

[Les]

Originally posted by unregistered


re the wb, all we need is a comparison - right ?
the LiquidCC Surge ?

Perhaps preferable,for cross-correlation purposes, a wb whose cooling has shown only a small sensisitivity to lateral movement*.This may minimise any effect of the difference in die size/geometry.
Maybe the Surge is such a wb.

* "I found, quite by accident, that the performance of some wbs is changed by mounting them 'off center' wrt the die....." - http://forum.oc-forums.com/vb/showth...0&pagenumber=5