TC-4 Water Block Changes Made.

[DodgeViper]

I made a change to the TC-4 with the turbulators. The stock turbulators restricted the water flow from my Eheim 1250. I had removed the turbulator, but now I made these little beauties. I have made another change from the time the photo was taken. Instead of having 6-7 coils per channel I now have 4 coils per channel. Having a round coil in a square opening I do believe will add turbulence. Water is going to flow between the top, bottom, sides, and center of the spiral coil, mixing with the water that is spiraling inside the block. It has yet been tested as I am waiting for a few parts to put my system back together.

[Penguin4x4]

"The CPU temps are reported from the sensor within the XP chip, not the motherboard sensor." In your sig..

How is that possible?

[The Spyder]

good use for a 14g or 12g copper wire!

[The Spyder]

Its called an ON DIE temp sensor......

[dxiw]

Originally posted by The Spyder
Its called an ONE DIE temp sensor......

lol but what about the TWO DIE temp sensor

[bongo]

DROOOOOOOOOOLING:clap: :clap:

I was thinking of doing a similair mod to my TC-4, but ended up having a COpper top soldered on. (It kept craking the poly on the dame spot). Can't really do this now.

But did you see any temp improvement? Do the coils provide enough surface area to create as much turbulance as the originals did?

[SemiCycle]

Pretty cool idea. I saw your post over at Procooling on this subject but never got a chance to reply. Are those coils spring like and "squished" in the channels, or are they custom cut to fit and don't act like a spring?

[Starfoxer]

wow! your temp is pretty damn low.

2.1volts and 1.93ghz brought me to 51C on my maze 2 + dtek 1/2 setup.

edit: BTW the block looks awesome!

[Iron Hawk]

I just put some spirals into my Maze3, so far temps are the same, bet i gotta wait till the AS3 sets before we can be sure.

[DodgeViper]

The coils were made to fit tight inside the channel. They do not work like a spring because they do not have any tension. I used #12 copper and I have made #14 copper spirals. They are cut to fit.

[nikhsub1]

Originally posted by DodgeViper
The coils were made to fit tight inside the channel. They do not work like a spring because they do not have any tension. I used #12 copper and I have made #14 copper spirals. They are cut to fit.
Beautiful creative work as always Dodge!

[Iron Hawk]

Originally posted by Iron Hawk
I just put some spirals into my Maze3, so far temps are the same, bet i gotta wait till the AS3 sets before we can be sure.

Ok, so far temps are up 2c, :( oh well, ill post pics tomorrow afternoon.

[nikhsub1]

Originally posted by DodgeViper
nikhsub1, your coils and Danny's are on there way. You should have them this week. I sent you #12 and #14 copper coils. Enjoy.
Thanks Dodge, you ROCK!

[nikhsub1]

My #14 springs are in! I am bleeding as I type this on one of my dedicated FOLDERS!!! I will have results in the next day or 2. For reference of TC-4 with NO Turbulators: XP 1600 @ 1871 Mhz, 2V with a system temp of 27C, I get 44C load (on die). Let's see what the springs can do!

[Lusankya]

Umm for watercooling those temps sound pretty high.. I currently get better off air.

Someone pelase tell me thats not normal for the TC-4 block cause thats the one I plan to use for my w/c setup

[nikhsub1]

Originally posted by Lusankya
Umm for watercooling those temps sound pretty high.. I currently get better off air.

Someone pelase tell me thats not normal for the TC-4 block cause thats the one I plan to use for my w/c setup
Umm, your MOBO reads the INSOCKET THERMISTOR, which is a probe in the ziff socket that reads the AIR TEMP, not the CPU temp and is VERY inaccurate. My mobo READS THE ON DIE SENSOR inside the core of the CPU, hence higher temps, but accurate. To give you an example, I have the same mobo (8K3A) as a dedicated folder running an XP 1600 @ 1.8G, 1.9V cooled with an AX-7 with a 50CFM Sunon fan. This air cooled machine loads at 54C with way less voltage and OC. You can't compare insocket readings to on die.

[The Overclocker]

Originally posted by nikhsub1
My #14 springs are in! I am bleeding as I type this on one of my dedicated FOLDERS!!!

ouch, i hope you whipe the blood of the computer - it really makes a mess

i cant see how turbulators can increase performance, they hinder flow - a open channel with little groves made in the sides like the becooling jagged edge would be better - hopefully these tests will prove me wrong

[nikhsub1]

Originally posted by The Overclocker


ouch, i hope you whipe the blood of the computer - it really makes a mess

i cant see how turbulators can increase performance, they hinder flow - a open channel with little groves made in the sides like the becooling jagged edge would be better - hopefully these tests will prove me wrong
Well the system is bled, but the AS3 will need some time to settle. So far, temps seem to be about the exact same, Case at 26C, Loaded CPU jumps between 42C - 44C. Rember, this is 2V, 1871 Mhz (XP 1600) using ON DIE READINGS.

[The Overclocker]

so... if the AS3 is not setteled, and you are getting the same temps as before - then you should see a drop in temps once it has settled?

[Lusankya]

Originally posted by nikhsub1

Umm, your MOBO reads the INSOCKET THERMISTOR, which is a probe in the ziff socket that reads the AIR TEMP, not the CPU temp and is VERY inaccurate. My mobo READS THE ON DIE SENSOR inside the core of the CPU, hence higher temps, but accurate. To give you an example, I have the same mobo (8K3A) as a dedicated folder running an XP 1600 @ 1.8G, 1.9V cooled with an AX-7 with a 50CFM Sunon fan. This air cooled machine loads at 54C with way less voltage and OC. You can't compare insocket readings to on die.

Where did you find this information? I have looked for it but never found it on either of soyo's sites

[nikhsub1]

Originally posted by Lusankya


Where did you find this information? I have looked for it but never found it on either of soyo's sites
I know that Soyo does not use the on die readings, this is what I found at their site:

Health Monitoring Description
On board voltage monitors
CPU fan speed monitor
CPU fan speed monitor
**CPU temperature monitoring through flexible thermal sensor** (ie, INSOCKET THERMISTOR)
CPU overheating protection

[Lusankya]

Hmm I'll have to look next time I pull out the cpu.. I remember not seeing anything inside the socket itself that resembled a thermistor

[JFettig]

Originally posted by Lusankya
Hmm I'll have to look next time I pull out the cpu.. I remember not seeing anything inside the socket itself that resembled a thermistor

its diff in the soyo boards, they have them anyways, and it doesnt make a diff its there or not, cuz itll be there no matter what, but it would say if it reads the on die diode

[DodgeViper]

i cant see how turbulators can increase performance, they hinder flow - a open channel with little groves made in the sides like the becooling jagged edge would be better - hopefully these tests will prove me wrong

I never made a reference that they (coils) would or would not lower temps. I got my best temps with the stock turbulators removed because I was using a 1250 pump. These copper coils have less resistance then the stock turbulators and might add enough turbulence to make a difference. This is why myself, nikhsub1, and Danny at D-tek are giving them a shot. Danny expressed interest first and foremost, as they are easier to make. Without giving it a try we would never know.

[nikhsub1]

UPDATE! With the #14 springs in, I am seeing good temps now, with my system temps at 25C, My load temps are 41C and holding. I will say this, they definately work better than the turbulators that came with the TC-4. I think I'll keep them in! Thanks Dodge! I would like to try some thinner ones, but these are rockin!

[nikhsub1]

Originally posted by DodgeViper
nikhsub1,

I could make #16 and #18 copper coils. Let me look into getting my hands on some wire. I think if I go any smaller the wire may give and not stay in place. This is good news. Post your temps again after the AS has a chance to settle in. I am still waiting some parts to do my testing, but we have basically the same set up.
I really dont think It will make a noticable diff, but would be interesting to see 16 & 18. When the system was bleeding, I could see major turbulence in the block, vortexes swirling, very cool idea you came up with Dodge!

[Since87]

Originally posted by DodgeViper
I could make #16 and #18 copper coils. Let me look into getting my hands on some wire. I think if I go any smaller the wire may give and not stay in place.

Might try silver soldering the wire to the walls of the WB in spots to hold it in place, and get some of the block heat into the wire for even better cooling. Of course there would be galvanic corrosion issues, but with the help of Water Wetter it might be worth it.

[nikhsub1]

Originally posted by Since87


Might try silver soldering the wire to the walls of the WB in spots to hold it in place, and get some of the block heat into the wire for even better cooling. Of course there would be galvanic corrosion issues, but with the help of Water Wetter it might be worth it.
Not needed at all, these springs fit TIGHT!! Not tight length wise, but tight in the width of the channel. They are VERY hard to get out. Dodge you did a super job! Temps now, case 25C, Load jumping between 40C - 41C. I think they should go down even a bit more as the AS3 settles.

[DodgeViper]

I agree there is no need to solder the coils in place. You take the chance of screwing up the block and once you place heat to the block it will discolor badly. The coils were made with a tight fit, so tight that you will need to make sure that they are sitting flat to the bottom so the Lexan top remains flat to the copper block.

Nikhsub1, this is really good news. I have a few other ideas I am kicking around.

[Penguin4x4]

what were your pre-mod temps, nikhsub?

[nikhsub1]

Originally posted by Penguin4x4
what were your pre-mod temps, nikhsub?
I am seeing a 1C-2C drop in temps with the springs as opposed to bare channels. I would say 3-4C drop from the stock turbulators.

[Penguin4x4]

Wow, you were runnin ~@50C with the Spir@l!!?:eek:

[DodgeViper]

Here is an updated photo of the block with #14 copper wire. Also the wire has only 4 coils per channel.

[JFettig]

cince you love to open that up so much, could you tell me the thickness of the base in there or dont you have the tools to measure?

thnx

oh you wouldnt mind if i stuck some of those in my s-type like just in the center, would you? or wouldnt you like it if i copied you

[Silver]

Bump, just by way of keeping track of the thread. Interesting read.:)

[JFettig]

i honestly dont see were the 10000dollars comes in, i hear my bro and my mom telling me i should pantent my block, obviously they dont know that its not worth doing, but were does the 10000dollars come in?

[R.Rabbit]

great idea! :D, nikishi, do you get better temps without the stock turbulaters than with?! thats odd!, do you think you could fit one of those spirals inside a spir@l?

[The Overclocker]

you could probably fit thouse 'springs' in any type of waterblock with channels.

dodge: do you think that the original 'turbulators' were just for marketing or that they were just badly applied?

[nikhsub1]

Originally posted by DodgeViper




I can answer all questions as I was the first to remove the stock tubulators. When using the 1250 pump we both (nikishi/DodgeViper) got better temps without the stock tubulators. Reason: to much flow from the pump and the stock tubulators caused restriction using a high flow pump.

Coils to fit the Spir@l W/B? Already made.

Personally I think if a person had a 1048 pump the stock tubulators would work fine.

Uhhh it's nikhsub1:D , anyway, I have the 1250 and the 1048 and with the stock turbulators, the 1048 does really well as Dodge said, with the 1250 it does not do as good. With the 1250, bare chanels or the springs work best, I see about a 1C better temp with the springs as opposed to bare channels.

[The Overclocker]

ok that has cleared that up , but finally does the tc4 do better with stock turbulators and the ehiem 1048 or the ehiem 1250 with the copper 'springs'

[nikhsub1]

Originally posted by The Overclocker
ok that has cleared that up , but finally does the tc4 do better with stock turbulators and the ehiem 1048 or the ehiem 1250 with the copper 'springs'
That is really gonna depend on the rest of your setup. For me, I think the 1250 bare/springs works a LITTLE better. I have not tried the springs/bare with the 1048 sooo...

[DodgeViper]

This post is for those that may make their own coils and replace the stock turbulators. Be very careful when tightening the machine screws. If over tighten you may crack the Lexan top. Only slight pressure is needed to seal the top.

[pauldriver]

Originally posted by maskedgeek
i honestly dont see were the 10000dollars comes in, i hear my bro and my mom telling me i should pantent my block, obviously they dont know that its not worth doing, but were does the 10000dollars come in?

Please be aware that Mini and Mainframe computers have been liquid cooling for 20 years now.

Any discussion that a patent attorney has with you about the patentability of a liquid cooling system is going to be disingenuous. He is simply stringing you along for patent fees.

Paul

[Penguin4x4]

Originally posted by nikhsub1


Uhhh it's nikhsub1:D ,

I like nikishisub1 myself,:beer: :D :D :D

DV, I heard(rumors) Danny's comin out with new TC-4's inspired by these new turbulators. Any truth?

[Overbrazil]

Oh god, have removed the turbulation too, and for my 540 gph pump the temp dropeed 5 C
I am current using pump - tc4 - rad - pump
is it better than pump - rad -tc4 - pump
wattage 100 W cpu
30 W pump (sub).
Thanks dodge and sorry my english

[Overbrazil]

Thnals, Dodge, i will try this order.

[DodgeViper]

I could not resist posting this photo of the new TC-4. Now this is one great looking block. The anodizing completely covered all exposed aluminum. The copper coils are installed. I will have my system back up tomorrow, but for a very short time. I am in the process of build a new Chevy Chevette Rad and my designed shroud. The new rad will be different than what is on the market today and unique. Also I will be using a TC-4 base with a three-barb top. D-Tek is not offering this block for sale, but an idea I am going to try in combination with my new rad. The block below is available though. Stay tune………

[SemiCycle]

Cool, it is nice to see that they are moving that block top to other models instead of just the spiral. I think it will help ease many peoples minds when making purchases. Just too much cracked top talk nowadays.

[DodgeViper]

The heater core with an additional barb.

[DodgeViper]

Drawing of my new setup.

[DodgeViper]

The Shroud.

[JFettig]

ok your drawing confuses me, it looks like yoru going in the outsides and out the inside? whats the deal with that? everyone beleves in it opposite of that, you got it done? or havent you gotten the top yet? post some pics of completion soon

[DodgeViper]

You’re not confused. I am bringing the water from the outside to the center. The coils are going to be made in reverse of each other so the water is turning in opposite directions.

[JFettig]

you may wanna try it the opposite direction just for fun, because you may never know, it may change things, i personally think those coils dont do too much.. maby you could try something: add a little die, and watch the water in your block, see if it does turn or just add turblence

[DodgeViper]

Having a dual output rad eliminates the Y connector and should have less backpressure. The center barb will be I hope 1/2" ID barb. 1/2" tubing between the pump and inlet and dual 3/8” ID tubing between the rad and block and 1/2" ID tubing between the center barb and return to pump.

Never been inclined to follow what everyone else does. The top will be made of either 1/2" Lexan or 1/4" Aluminum, but anodized.

[DodgeViper]

I thought I would revisit this thread and bring back what this thread was about at the beginning. My system is back up and running, and I have some results with the coils installed. After the house temp was stablized I can say that the coils do help and they will remain in my system. I gained 1c lower temps over an open channel TC-4. The gain over the stock turbulators is 3c. These results are the same as nikhsub1 had reported after he had installed the coils. These figures are while using a Eheim 1250 pump and in my system and may/will vary from system to system.

Once I get the new system online as described above with the new heater cord and new top on the TC-4 I will post the results on that setup.

[R.Rabbit]

hey dodge i know you obviosly have contacts with d-tek, so when can we expect to see the new tc-4's with the anodized al top and the spiral thingies inside? im wonderin' because its the only part of my rig thats not on the way. :D:D

[grim]

go check the site. they are available for order

[R.Rabbit]

oh! nevermind! thanx dodge!
(i've gotta read my messages more!)

[pauldriver]

Does anyone know if the DTEK Vid-jacket will fit a GForce4?

[nikhsub1]

Originally posted by DodgeViper
I thought I would revisit this thread and bring back what this thread was about at the beginning. My system is back up and running, and I have some results with the coils installed. After the house temp was stablized I can say that the coils do help and they will remain in my system. I gained 1c lower temps over an open channel TC-4. The gain over the stock turbulators is 3c. These results are the same as nikhsub1 had reported after he had installed the coils. These figures are while using a Eheim 1250 pump and in my system and may/will vary from system to system.

Once I get the new system online as described above with the new heater cord and new top on the TC-4 I will post the results on that setup.
This temp drop should be noted that this is with the Eheim 1250. With the ANY OTHER PUMP, this may not be the case. Just want to clarify.