Well after owning this Mach II GT for a few months now, ive finally got around to switching it on for the first time.

Its a Lian Li pc 60 top with matching base unit. Firstly i removed the cover to the condensor. The dust, hair and god knows what was about an inch thick! So thats all removed now. Cant believe any air could have possibly got through that lot

Fired it up and it became apparant that the temp sensor is waay off. At room temps it reported -10C and when it had fully got down to its max cooling under no load, it reported -80C Is that just a dodgy temp sensor or dodgy controller board? hope its just the sensor. Anyway the head frosted up nicely

Got sick of waiting for the crossfire boards and with this fx55 SD sat doing nothing so got a Neo2 on its way to me. Should have it all installed for the LAN hopefully or i wont have a rig to use. Only got a 9800pro to use in it so no big benchie results just yet

Pics to follow when i get some batteries for the cam

Righty then.... temp sensor. That sounds like a unit either I've repaired, or have acted as advisor on the method for the repair.

Basics being, when the temp sensor popped, generally the unit wouldn't read cold enough. There was a solution for this. Make it read colder than it should. Easy way - take a bog standard temp sensor and attach (ie: budget temp probe stylee). So yes, the reading is way off but it's enough to get it into the minus temps and convince the controller that nothing is wrong.

Few things you can do to try and get it back inline. Thermisters work on resistance. Add a variable resistor onto one wire of the probe and twiddle whilst watching temps... find out what a stock GT generally runs at when not connected to CPU after xx amount of time. Set yours running under same conditions (same ambient if poss) and after the same xx amount of time, twiddle that resistor til your temp matches.

If ambient differs, then for every 1 deg change in ambient, effect a 2 deg increase in evap temps in the same direction.

Thanks for that. It was Harry's old unit. Left it running for 20 mins and came back to find a nice lump of frost on the head

New mobos coming tomorrow, then its time to get the FX under the freezer. Will get some batterys and take some pics

Installed the mobo in the prommi today. Now i know why ive been putting it off for the last few months, anyways its finally done.

Havent got a HDD yet but boooted up into bios to check it was all working. Got a cpu temp reading in bios (yes i know it means nothing) of 37C at stock speeds 200 x 13, raised the multi upto 14, then 15 and then 16 (all on stock volts) and it sits in bios at 3200 with 1.42v. Temps in bios go upto 40C.

When i get a HDD sorted, then install windows, i should be able to get a decent beta bios to allow more volts

Also want a few more washers under the mounting screws cos im not satisfied that i have good enough contact.

With a bit of luck, it should all be installed for the LAN, if not i will be a spectator with a case of beer

Is this normal cos im not sure.

The evap head is actually warm and it you touch the head mounting screws they feel real warm (ok they have sealing string on them) I dont think it is quite right

The last Mach i had, the head was real cold and under no load you could get condensation appearing if it was left idling for long periods.

The evap readin on the Mach is -76c which i know is miles out. In bios its reporting 40c idle too

Yep, the evap. head is warm because of the heaters in them - to reduce the chances of condensation forming!

Strange how the old one didnt though unless the heaters weren't working maybe.

Anyway it seems pretty stable which is the main thing. Currently running F@H for the last 3 hours at 3.2ghz. Will hitch it up to 3.3ghz later when i get in and leave it all night to see how it goes

If controller is set to Turbo then heaters turn off. If controller is set to normal then heaters are turned on, which produces a heated head and screws.



Now that ain't good. Should either be reading as 24x deg C or 0 deg C or a - figure. Certainly shouldn't be giving 30's / 40's readings.

S'either a TERRIBLE mount (unlikely) or the unit has a more serious issue.

Whack fans on turbo, thus killing heaters and leave it to idle. See if you get your condensation / cold head feeling back again...

Need to find out exactly what that CPUs running at (ballpark)... confirm with other Neo2 owners what temps they see when running phasechange - ie: can board read - temps or not etc etc.

Next tip to test whether mounting - move unit up side of a wall so mobotray side of case is up against the wall. Take side panel off. Grab evap head... lean on it so your pushing head down onto CPU, with wall on opposite side to give u summat to go against / stop rig falling over. See if CPU temp decreases.

If you have any waterblocks knocking about, nylon washers off the mounting kits - use those.

Well the mobo or the chip has died

Hot flashed the bios and still wont boot up. Displaying all the signs of a dead bios (fans spinning on power up but nothing)

Just before it died it kept restarting when trying to boot. It would boot into safe mode only so tried a fresh install but kept rebooting during the install. Stuck everything to stock settings but still the same. Just stripping the mobo out now to check for possible condensation probs. Will report back later

I was running it on turbo and got no condensation or cold head feeling. It was completely stable at 3.2ghz 100% for the last 12 hours.

I checked the mount too and that was ok

Prommi's futzed d00d. Get it to the shop next week and I'll cast me eye over it...

So, it freezes up nicely when not connected to PC... but with load performs like a bag-o-poo. Generally indicative of either fans not working on evaporator or refrigerant leak. I'll assume you've already had your head inside to make sure BOTH fans are going... so I'll in turn also assume it's got a leak somewhere.

Cheers for that.

Just removed mobo n cleaned all the string off. No bent pins on the chip, gonna whack a heatsink on it and pray

Quick n easy leack check, run Prommi on it's own, work your way round the tubing with yer ear and see if you can here a typical air-escaping noise at any particular point. If you find such point, slice insulation with craft knife and roll back. Spray some soapy water on, see if it blows bubbles or sucks it in. If it sucks it in, turn off and leave it off until repair can be done (cos it's just sucked in a load of water which will need purging unless it's drier is in good nick). If it blows bubbles then leak is on high side and is blowing out.

W00T!!! it booted on air!

chip n mobo not dead

but your mach2 is. we have a certain member that will fix it for you.

-76 sounds awful cold for a standard GT. maybe for a 507'd GT but not on 404.

The 40c mobo reading is normal if you're using an older bios for your Neo2. The newer bios's will report down to around -13 (like mine does) but not any lower.

If it's saying -76 it's either some kind of probe problem, or it's got a little leak and it's starting to get colder. Colder, but less wattage capability.

Run it on it's own. Does it seem to take longer to get down to temp?

When you had it on there, going by the display, could it seem to hold up to the load?

And yeah, if you like, I'm more than happy to help as well. If Marci has a look and figures it's in need of leaktesting and yadda yadda, I'm here for ya. I just got in a load of stuf from the supplier, and my flu is finally starting to calm down a little. Well, I can eat now anyway.


Gray

Yeah the bios used was the stock bios that the mobo came with.

It got down to to temp real quick, after about 30 secs it was dropping around 1C every second. It held temps ok too. At the max i ran it (3.2ghz) the display went from -76ish to -71C with 1.55v to the core.

Thanks for the offer Gray, im sure we will be meeting up pretty soon

Right it think the mobo is not happy. I can boot into bios but as soon as i save the settings to make it boot from the cd, it appears to save and reboot as normal but keeps rebooting and doesnt get into bios at all. Only way to correct this is to clear cmos

This is running on air now

Its trying to install windows but reboots during the setup. Now ive changed the ram, underclocked the cpu from 2600mhz to 2000mhz, still getting the same

Just got a IRQL_NOT_LESS_OR_EQUAL message with STOP: 0X0000000A (0X00000008, 0X00000002, 0X00000001, 0X807CAA66) message

Well, that can only be a driver issue of some sort... am assuming all you have in is CPU, RAM, GPU, MOBO, HDD, and Optical Drive...?

If optical drive is DVD, change for non-DVD drive (ie: CD or CD-RW)
If HDD is SATA, head for different drivers for SATA Controller... check a few older revisions / ensure revision you're using matches bios you're using.

Update mobo bios via old bootdisk method if can't get windows on... seem to recall a few folks having issues with NEO2 and stock bios when it came to installing windows...

Hell, if yer bored, rag the rig over now and I'll have a look - got plenty of alternative bits to swap in and out here... just don't leave it too late...

Oh aye, and set command rate to 2T and cas timings to bySPD if u haven't already... restore CPU to stock clocks... s'not a processor issue, am fairly convinced of that...

Handy reference for MS STOP messages - http://aumha.org/win5/kbestop.php

Cant get over to the shop today, no transport

Tried changing the ram (3 different sets) Changed the HDD (from SATA to IDE) still getting bios freezes. It will only go past the bios if i completely switch the power off, leave for a few secs then back on. It freezes on the default bios pic if you just reset.

The windows install i had working on there yesterday, if i hit safe mode to almost boots but restarts. Its like the psu isnt powerful enough to boot up. Rails look fine in bios, will have a looksie through my bits n bobs n see if i have a spare.

Only things connected are GFX, HDD, CPU, RAM and DVD drive. Tried disconnecting the DVD drive but getting the same probs

Only thing left for me to try is another PSU and failing that, the mobo is the last

With a different PSU still doing the same

gutted dave, if u need a spare 939 mobo ive got an asus a8v spare if thats any good, if u still need a socket a cpu ive got an xp1700 hanging around too

Well from the sound of the prommie, the temp reading is a little low, but nothing serious wrong unless you like a good temp reading.

Really sounds like a mobo that needs to 'dry out' for a while.

If you've got some parts cleaner, try to get any and all moisture outta that socket, cos I'm really suspicious of it being that.

Did you use di-electric?

I've had all sorts of moisture problems and this really has the ring of that. If you have it 'loose' I'd say get it onto the oven and get it nice and warm and see if it can speed up any removal of moisture if that's what's going on.

Gray

I used vaseline cos i had no grease at hand. Think it was safe to do so, but didnt think that the board might need to dry out.

When it first started going wrong, i was getting random reboots which as you know can be a sign of condensation so i removed the head and had a peek. Couldnt see any but there may have been some traces of moisture that i couldnt see. Ive fitted a NF7 in there for now and fingers crossed it should see me through the LAN without problems, that should then give the Neo2 time to dry out and i'll give it another go before trying an RMA.

In fact the missus has just put the rads on so i'll stick it next to that for a few hours

Any time I get 'strange problems' when running phase it's usually water ingress. you've got to make sure all your clocks balance of course, that it vcore to cpu clocks, vdimm to timings to dividers etc etc, but if that's all ok, then that's what I usually look for.

I've never used vaseline but I've seen it used with no problems.

I really hope it's not the Neo2 disease. Best clocking A64 AGP mobo, but seems like it's always had issues when you freeze it. I've been lucky so far myself, but I've stil had one dead one.

I hope it's ok

Gray

OK - now you've got my attention - I'm getting exactly the same symptoms as you and even started getting the same error message (Reboots while loading windows etc...) Got to the stage where it would boot up and the Splash screen was just random colors, like a completely corrupted loading screen - thought it was the bios, but after leaving it alone over-night it magically cleared up in the morning and let me back into the bios - was about to try re-installing windows, but your experience makes me think it's something else (water?)

Running XP-M, NF7-s ver 2, various memory, IDE HDD, CD-Rom, DVD-Rom. Got a Mach 1 on the chip, and a little R12 chiller cooling 172W pelt.

Keep us updated on your progress please.

Certainly will. Got the mobo leant against the radiator for now. You know when i moved it hours later, it may sound strange but the back of the board was ice cold (well the metal bracket used for holding down the air cooler. There was a mild film of condensation too. Its not like this room is extremely hot or cold either. Just hope the mobo aint dead. It was running peachy for 12 hours at least

why not try one of the agp dfi boards dave its gott be better than the neo 2 , i hope

Yeah, suppose so but DFI and me dont seem to get on unless the newer boards are better

Right going to the LAN for the weekend gave this Neo2 some time to fully dry out. Its been resting against the radiator all weekend. Had an hour spare so hooked it up on the desk in front of me. Fired it up into bios. The last time i tried this it would get into bios for a minute max, then either lock up or reboot. Now its sitting in bios on a stock A64 cooler (not the FX air cooler) running at 2.8ghz with 1.5v. Its saving the settings im throwing at it unlike before so it seems that the board has fully dried out and appears to be ok. Right dont know if the cooler was ok all along cos it held 3.2ghz for around 12-14 hours F@H stable. Maybe the probs ive had are either a not perfect mount and let some condensation in or the Neo2 doesnt like the cold too much, maybe a bit of both.

I didnt see any condensation or water in the socket area when i stripped it all down but i suppose it would only take a small drop to cause the probs i was having

Im just glad that everything appears to be working ok and it lived to clock another day

Righty then - u know the best thing to do already - run it on air for a few days with a windows install, make sure it's folding stable etc and test all the components out thoroughly on air. Then you know they're not the cause and can firmly ruleit down to Prommi / Installation thereof when the time comes.

In the meantime, prep it up with the relevant bios' suited to low-temp running if you can find one...

Always wondered tho... the hole in the IHS of CPUs... airtrap = condensation under IHS risk? I always try and fill it in with arctic ceramique irrelevant of what paste I'm using for the actual TIM joint...

Never known whether it was an issue or not but always done the above as a precautionary measure...

That may be an issue cos i know the IHS was removed before i bought it. It is actually a little loose but i didnt think that would be an issue. I think i'll carefully remove it and get some ceramique in there. Also the mounting kit appears to of been modded to fit Intel mobos too. Thought i'd sealed the extra holes but that might be another thing to recheck.

Is the IHS actually loose Dave? When I put it back on I was pretty careful about clamping it right. I DID use AS5 under the IHS though, so if you want Ceramique you'll have to r&r the IHS.

I left the airgap as it was, for the main reasons that 1. getting it to seal would be a pain without air for the sealer to cure on the inside and when you clamp it down, there's gonna be a little pressure inside under the IHS pushing it up, and I didn't want to risk not getting good contact. 2. if there's moisture in there at all, that airgap is the way out.

I'd only recommend sealing it after you've removed, cleaned and repasted, refit and allowed to cure, and THEN seal the airgap with more sealer.

When i refit the IHS on that one, I clamped it with the tightest of the 2 aftermarket heatsinks I have (TT Tower) and allowed it 2 hours for initial drying time for the sealer. Then I powered it up, and let it idle, so the paste would settle out before the sealer was fully cured so the paste and sealer would both compress fully. Did that over the next 3-4 hours, on for 10 to 15 mins then off for 45 mins. Speeded up the paste's break-in, fully compressed the sealer.

Temps dropped about 8c or so over that and then after it was left overnight I couldn't budge the IHS so the Sealer was 100%. Only thing I can think of is if the Sealer got loosened by the Xtreme cold.

Sorry if it's letting you down. If you want me to do the repair or refund you just let me know.

Gray

Hey you didnt let me down at all Gray. The IHS lifted on one corner the first time i'd had it on the air cooler. As i revomed the cooler, the chip came with it, thats when i noticed one corner was kinda loose.

Just trying to install windows onto a spare drive i have but getting similar blue screens as i was before. Just trying to find out why

Disabled the follwing and it appears to be installing

• L2 cache
• BIOS cache
• Internal/External cache

Hmmmm.... bios cacheing you want disabled anyways, but L2 cache and internal/external you REALLY need enabled... you're gonna be dropping down to Pentium2 MMX performance levels...

Soon as windows is installed, re-enable them... then if errors start showing up, downclock the CPU gradually. See if you find a point where the errors stop...