Wow I made that post months ago. Took me a while, spare time was a bit rare.
Currently planning to go cold for the first time this weekend. Painted a trusty old P5Q-E with the liquid rubber.
I used the clear one because I really loved the idea to see through and it helped me to learn how to apply it correctly. Three coats of rather thin layers, carefully spread out with a brush. Otherwise you get airbubbles when it dries and shrinks up.
I also made one big oopsie:
No idea how it happened but some of the rubber got into the socket... There was no drop on top of the CPU that I kept in there to protect it from drops, it must have crept in from underneath. I thought I was careful enough and left enough space around the socket but nope.
The red marking shows the area that was affected, nearly flooded half of the socket with the stuff. This gave me a small heart-attack when I saw it but its actually no big deal. I let it dry, put some alcohol on a 2000 grit sandpaper and gently brushed the pins 3-4 times.
Was another of those moments where you are super anxious pressing the power button but it just fired right up.
Gave it a testrun with just water to check on the VRM temps, which are now coated. I thought it would be really bad but its fine. These dual cores are rather tame in terms of power and the VRM is big enough.
Man am I glad its running. Everything down to the PCIe slot is covered back and front on the board. Just taped off the connectors and DIMM slots.
I also got some armaflex and shop-towel ready.
I'll start with dry ice this weekend and see how it goes. If my method works and everything survives the weekend (I doubt it) I might jump to LN2 but for now I think dice is the easiest way to get my feet wet.
Post by Macsbeach98 on Jun 25, 2020 18:46:05 GMT -5
You will do fine I know what its like getting the paint in the socket it wicks up from underneath beware of it doing that in ram slots too. In my early days I got red LET into the socket of a P5Q3 Deluxe Wifi so I cleaned it off the pins with Acetone and a soft brush after it ran it cold and pulled it apart it was all over the bottom of the CPU again it was like that for a few runs after it happened I still have the board I am sure if I was to run it now it would still come up in there.
Here is what I am doing now for insulation LET put on with a syringe its quick and easy in the second pic you can see the way I have my foam pads so it seals off the CPU completely that was a Prescott 478 in a 775 adaptor had it running for over 2 hrs at -100 no condensation problems at all. 775 CPUs a good one will run comfortably at -120 a great one will run at -130 my QX9650 locks at -90 they dont go to full pot more insulation is needed on boards that do. Your insulation methods will change with experience as you work out what works for you.
Applying my crypt(ic) skills here to zombify this thread yet it's relevant.
Here's a guide for those of you that may be wondering how I did the stalk mod to my pot. I may need to revise it later for corrections but for now this is it and how I did it. This info is useful for other things you may want to try as well. -----------------------------------------------------------------------------------------
Copper normally is hard to work with concerning things heat related because it passes heat energy with ease, that's one reason why you can't just cut it with a torch......
Unless you know HOW to stop it from passing heat along to another object/surface so readily and that's key to what I'm about to tell you.
This is just being used as an example of how I work with copper but it is based on real work I've done and you guys have seen.
For this you need a few things prepped in advance - At least if attempting to do the mod I did.
Remember safety first - Because I have to say it for good reason.
Make sure any and all combustible stuffs are away and any flammable gasses/fumes are not around or the potential of it being released when doing the work is removed.
For the purpose of doing the mod as I've done with mine, you need a "Place" you can set the pot on that won't burn or catch fire, the pot and this place will both be hot and exposed to an open flame while the work is being done so that's important.
My place is a workbench that has a bench vice I use to hold it.
Although my bench is made from wood I can still do it because the flame never gets close enough to light it, a metal surfaced table is best however for obvious reason. It's best if the plate isn't directly over anything so you can apply the torch with ease and not to light anything just because the torch is there.
*Now it's time to get into things about it.*
To start off I created a plate for this work.
Many think of it as a hotplate and that's accurate enough because that's really what you'd have and be using - Just that it's not an actual hotplate as you think of it being.
I used a piece of flat steel plate for my setup of 1/8 inch thickness so it's strong enough to hold the weight of the pot yet not so thick it's a pain to work with because of it's weight. I also had to be sure it wasn't bent or warped so it would have a consistently flat surface - That being important.
I made this as a square shaped piece about 4 - 4 1/2 inches square all around with a small "Leg" on one corner that was bent down 90 degrees from it's surface. This leg can be made either by simply using a hacksaw on one corner and doing a simple cut of about an inch and then taking this and bending it down.Be sure to make your cut about 1/4 inch from the edge so this leg once made is strong enough to hold the weight without bending under the weight of itself and the work you have on it.
You can also just get a small piece tack-welded in place too but I prefer the hacksaw method because there is no weld to possibly break due to what you have to do for getting things level for the work - It does involve bending this leg a little at times to level the work and that in turn will weaken a welded joint so that's why.
Be sure this plate is smooth as in sanded smooth since you will be placing the pot's finished surface on it and always check it for any signs of roughness/rust spotting or whatever each and everytime you use it. NEVER slide the pot to move it of course, always set it down as in straight down on the surface and pick it up the same way everytime.
I'll now explain how this works: The reason this works is because with the small/thin leg, you're reduced contact with what's holding the plate the pot is sitting on, reducing the area heat can transfer away from the plate and that means it can get hotter and stay hot for a longer period of time.
Simply put, the plate itself gets hot and then the pot itself heats up and holds enough heat long enough to do the work. No need to worry about the pot itself picking up heat from the plate because it will, it's just that it can't pass this heat along to something else via the plate too easily so it gets and stays hot long enough for the work to be done.
Now, once you have this plate made and ready, you just place the pot on the surface and breakout the torch. The plate when I mount it in the vice sticks out from the table due to how I position the vice and plate when I clamp the plate's leg in so the work is closer to you and easier to use the torch on too.
I never torch the solder/pot directly for a couple of reasons.
Even though the torch won't hurt the pot it's best done by NOT applying heat directly to the spot to be soldered. The left-over gasses/byproducts of the gas burning can contaminate the solder, weakening the bond itself - That's why indirect heating is how I do it whenever possible.
This doesn't mean heat applied directly to the pot would actually hurt it.
This only means in cases you need the pot/piece held where it is for the work to be done while miminizing heat loss for the work to be done. In this case the stalk is down inside the pot itself so you can't have the pot upside down and have it to work because it won't - The pot must be upright and level at all times during the work. You could try holding your torch nozzle down into the pot but that doesn't work well either for several reasons, including about the solder being subject to contamination, weakening the solder/bond itself.
With the torch I hold it to the plate's surface underneath the pot and let the plate itself get hot so the heat rises to the pot through the steel. Once the plate gets hot enough the pot will too and that's when you can solder whatever to it with no need to worry about a soldering pen or iron in use. The plate being what it is will retain heat long enough to let you work with it for a couple of minutes (About 90 seconds) and that's probrably the biggest thing about it, you don't have to worry about reheating the plate every 10 seconds and the mass of the pot itself will retain heat for a little while itself.
As for prepping the pot for the stalk: What I did was first, I made sure the pot itself was as level as I could get it, that's when you may have to bend the plate around on it's leg a little but it all must be level for the next step or else. I "Seeded" the spot for the stalk in the pot by adding some solder by snipping off small pieces from the solder roll and just dropped them in the pot where the stalk would go. Once I was sure I had enough in, I got it all hot so these pieces melted, settled flat and fully to create a flat surface inside where it's to be, then I let the pot cool. It's important all the solder melts and has a chance to flatten out level before it cools down.
I made sure the entire spot down in the pot was covered with solder so I'd know there was enough in place. Before moving on you can add more if you need to but you really don't need too much, more or less enough to completely cover the bottom of where the stalk will be.
NOTE: *Solder WILL want to "Climb" or wick it's way up the stalk, too much in the spot and you won't be able to spread out the tips at the top to aid in heat transfer to the LN2 once all the rest is done.*
Next, it's on to making and prepping the stalk itself.
I took some spare copper wire I had of the thickness needed for the stalk and first, measured and cut it to the length I wanted. Next, I made sure the ends were flat by letting my benchgrinder do that for me so when done it could stand on it's own and not wobble, plus I used a fine bastard file to complete the smoothing of these ends - This is important to do.
I test the piece by trying to stand it up on my bench, checking for flatness on the end you want to go down when you set it in the pot. I did all this before removing the wire's insulative sheathing.
After that I used a razor knife to begin removing the sheathing by first, placing this piece of wire on the table right on it's side and "Rolled" the wire under the razor to create a slice all the way around the wire, right in the middle of it.
Next, I then made a slice on one side of the cut from center cut to the end that would let me peel the sheathing off by just standing the piece up. I held it and sliced it straight down the side so it could be opened like a banana and just peeled off. Before slicing the other side off I used a couple of small squeeze clamps on the bare wire itself to hold the strands together so they woudn't come apart, then did the other side and used the same kind of clamps there as well. You also want to "Balance" how you use these clamps, simply placing them all on the stalk the same way will tend to make it heavy to one side, when you do the following balance test be sure to have it right so it still stands on it's end while on the bench.
Beware the individual strands of the stalk "Falling out" of the clamps because they can and at least will try to in most cases. The clamps will hold them but at the same time there will be a slight gap in contact with the clamps and any strands in that spot will move. I normally counter this by placing the clamps on so there is at least some contact all the way around the stalk either at the top or bottom of it and I handle it carefully anytime I pick it up or just move it.
Next, once the stalk is ready double-check things with the pot's leveling and you're ready to drop the stalk in.
Now at this point the wire piece is ready and the solder is also in place down at the bottom of the pot with a level surface and this is why having the ends of the wire piece flat is very important. Before setting the stalk down in the pot I tested the stalk to see if it would stand on it's own on a flat surface and if it did, I moved on but if not I did what I had to to acheive this.
Next, I applied some flux to the flattened end of the stalk and then set it down in place with this flux'ed end down towards the solder spot I made and double-checked to see if the wire would stand on it's own once it was in place to be soldered.
Note at this point you can gently tap the very top of the copper strands while it's sitting in place to make it stand upright if you must but only do that if you have to - I made sure the strand ends were even before setting it into the pot so hopefully I didn't have to disturb the stalk "As Is" once it was in the pot and standing in place. It's possible the strands could move/shift around while handing it so if you have to, work with it until the stalk is standing up straight before applying heat to the plate/pot. I used a long pair of needle nosed pliers to carefully set it down in the bottom of the pot and went from there. I grabbed one of the very ends of the top clamp to do this - NEVER try it by grabbing a strand, it will move and possibly pull out which will make you have to completely redo the stalk (Make it again) so don't do that.
If it did stand on it's own down in the pot and was standing straight up as it needs to be, that's when I took the torch and got the plate and pot hot again.
*Caution*: Do not bump the pot or cause anything that would make the stalk fall over during this time such as bumping the pot, bench, plate or whatever else, if it does fall over you'll have to make it stand up again or even remove it while the pot is hot to reset and try it again.
If all goes well once the pot gets hot you'll see the piece of wire "Set down" on it's own without falling over to one side and once it does settle correctly (As in standing straight up), remove the heat and let all cool.
After the plate and pot have cooled you can then remove the pot from the plate for the next step of clamp removal. I set the pot down on a soft surface and used a thin pair of needle nosed pliers to reach in and squeeze the tips of the clamps to slip them up and out from the stalk.
After all the clamps were removed that's when I took a small screwdriver and carefully spread the tips of the wire strands to how they are now so Ln2 can get down a little into the stalk for better cooling and surface area. Just bend these tips out according to how the strands are wound, simply pushing/pulling them straight out won't work but simply spreading them out along the curvature of how they were wound together is how you do it. Aside from cleanup and testing - Done.
As for how I did it, that's the basics of how you solder copper without a soldering pen or iron whether you are modding a pot or not.
austin86: Someone is selling a open box DFI lanparty 875p board on ebay ebay.us/nFs7FT and one way over priced 462 lanparty board lol ebay.us/5glulB
Sept 25, 2023 12:25:47 GMT -5
austin86: antinomy Ram and CPU are good, the problem I keep running into is the hdd not being accessible. I was able to work around this by booting from a floppy and I can run Pcplayer and Speedsys within dos so I'm guessing the system is stableish.
Sept 25, 2023 8:21:32 GMT -5
antinomy: I think you should start a topic on this one. It could be several reasons you need to check - memory, Vio, badly clocking CPU.
Sept 20, 2023 18:18:11 GMT -5
austin86: Stuck in the low/mid 170s and the system is stablish when booted of a floppy to about 180. IDK maybe I need to just toss deice/ln2 at it.
Sept 18, 2023 13:50:45 GMT -5
austin86: Does anyone know of a ISA VGA/IDE card? trying to hit a record on FSB with 440bx and I keep getting held back by the on board IDE, I tried a ide to sada adapter and I got further but not by much.
Sept 18, 2023 13:49:43 GMT -5
antinomy: What test? It could be because of AGESA version or maybe some security patches for the CPU.
Sept 14, 2023 0:48:33 GMT -5
george: And it was not a temp issue as far I could see.
Sept 13, 2023 18:44:43 GMT -5
george: wth? Updated to latest BIOS version on the A520, did a few bench with cpu-z after that, hmm, results lowered, anyhow got the Ram settled ok. However, increased the CPU some 5MHz, bench result went down? Perhaps forgot some settings I had.(This W10)
Sept 13, 2023 18:43:16 GMT -5