I just did a rough flow test.

This is a 200gph submersible pump with unknown other factory specs.

Pump alone at 1ft: 1 gallon in 16 seconds.

Pump with KingPin: 1gallon in 32 seconds.

Note this is the bucket test. Done it 5 times and averaged the results.

Next I will test the Maze 4 and my Lemon Block Cu.

Lemon Block Cu: 24 seconds per gallon.

Maze 4: 29 seconds per gallon.

KingPin: 32 seconds per gallon.

Tomorrow I should have the test bench setup well enough to give me a rough idea how well this block will work. I will run this block, the Maze 4, and my Lemon Block Cu and see what happens. I expect the lemon Block Cu to outperform both and even expect the Maze 4 to beat out the King Pin. Just have to pick up a couple hose clamps and the test bench will be operational.

This is what I have so far on my test bench.
http://www.customcomp.us/testbench/001.jpg
http://www.customcomp.us/testbench/002.jpg
http://www.customcomp.us/testbench/003.jpg
http://www.customcomp.us/testbench/004.jpg
http://www.customcomp.us/testbench/005.jpg

I still havn't got all the probes wired up and in place but I do have the DMM taking the core temp under the CPU with the socket mod. I also have a probe I will drop down the filler for now to take water temp.

I am not going to get to technical with this bench. Just using ot to test my own blocks.

The block is now running a TBred A core 1700+ @1800mhz 1.8V. Been running for an hour now and waiting for equilibrium to set in. Not going to do a hard core test. I will watch it for a few hours and take some numbers then up the Vcore and MHZ and run it for another hour or so.

That will give me a "good enough" idea of it's performance. I then will drop the Maze 4 on and do a rough comparison.

I'm wondering Jaydee if you used the roughing end mill on the Kingpin?

The irreguler surface of the pins done with a roughing mill would tend to reduce flow, but would also offer more surface for better cooling.

Be interested in seeing the outcome regardless. And while your test bench isn't all you'd want for doing block testing to post reviews with, it's still better than what the vast majority are useing that ARE posting review articles now on the web! And it'll give you a start on doing comparison testing. That experiance may help you choose further improvements for your test bench.

Luck with the testing.

I can't find one small enough. Everywhere I looked the smallest one is 1/4". I seen a couple 1/8" but they don't have the right edges on them. I am still looking though.

My test bench failed me last night. One of the O-rings in the impellor housing on the pump decided to leak. Lost half the coolant (50/50 water anti freeze) all over the table top and the computer was shut off. Hopefully the overheat protection worked and shut it off. I didn't have enough time to check into this morning before I went to work. If it isn't fixable then I am down and out for a while. Don't have the cash right now to buy a new pump capable of using for testing. Will have to wait untill after New Years. If nothing else though this will give me time to get my thermal probes sorted out and calibrated. I spent the last 3 days testing that block and got no solid temps. I think the TC on the DMM might be wacked. :( Lot of work to do...

Sorry to hear about the pump Jaydee, perhaps you can replace the seal?

Glad it didn't cause damage to other parts of your set up when it went out like that.

I have a disturbing feeling the housing is cracked. The pump is not that old and the o-rings should be fine. I just don;t see why it would leak...

I designed the test bench to keep the coolant off the white carpet if it ever sprung a leak and it succeded! Not a drop made to the carpet. Hopfully everything is in tact. I didn't have time to look it over this morning. First thing on my to do list when I get home. :D

Continue the good work.
Finally we may see some more serious WB testing.
I missed that.

Found 2 cracks, one is the problem. A freaking crack right in the O-ring groove. Second one is at the inlet but but it isn't leaking. Now to figure out WHY it cracked....
http://www.customcomp.us/pump/003.jpg
http://www.customcomp.us/pump/004.jpg
http://www.customcomp.us/pump/005.jpg
Bah... Hopefully the ISP Weldon #4 will fix it...

Well after a little ISP Weld On #4 it seems to be ok. Been running full speed for a couple hours now.

Took my TC out from under the CPU and took my indoor/outdoor probe and unslodered the thermistor from inside the housing and attached a set of wires to it and soldered the thermistor to those wires so now both the probes are able to be calibrated. So now I am calibrating them and as I suspected they arte all messed up. So I am going to try the joy stick mod. I should have enough thermisters laying around and plenty of old busted joysticks. :D
http://www.benchtest.com/gp_Temp.html

What can a pump manufacturer do to minimize the heat from transfering into the water?

Make the impellor and the housing not metal and make it not attached to the steel shaft. Study some pumps it is pretty basic. The Danner and VIA's and similar are all compacted together with the impellor directly on the shaft. Pumps like the Little Giant previously mentioned and my HydroThruster500 have true mag drive impellors. That means the impellor is NOT attached to the shaft of the pump. The housing is plastic and moved away from the pump motor so the heat doesn't transfer to the water as well. Iwaki is another good one as they take extra efforts to keep the heat out of the water. it is very important in some aquariums to keep as much heat out of the water as possible for some of the more touchy fish.

As I understand it, only a small percentage of heat is given off at the motor end (5%). What you described should make very little difference to the total amount of heat being put into the water. Maybe 1% at best. This would be nowhere near enough to compensate for the different wattages between the little giant and the two smaller pumps. It's because of the saltwater enviroment that the pump was designed for that the propeller was as you described . Heat had nothing to do with it.

where can one purchase one of these little giants ???

All I can say is you need to find a new information source. 5%? Where is the rest of the 95% going? :rolleyes:

Here' the link to the thread where I based my argument from

http://forums.procooling.com/vbb/sho...t&pagenumber=1

I reread the thread and I realize the motor heat is more like 20%, not 5%.

I will buy that. It really depends on the pump though. Especially if it is submerged or not aswell.

I left my Little Giant 170GPH (similar to a Danner Mag pump) runing inline without the fan on on the rad and the water heated up 20C over ambient. Thats just the pump running and no CPU or any other heat source. Some of that is the friction of the water flow.

My HydroThruster 500 with the pics in the last few posts will add 15C to the water with almost 4 times the flow rate. It adds less heat because of the design.

Anyway enough about pumps in this thread. Not what I started it for. :D

Any chance to do additional tests on the Lemon Cu, or Kingpin with the repaired pump yet Jaydee?

I'm glad to hear you were able to effect repairs on the pump. But will not the sealed crack still be prone to extending the crack with the vibration of the pump over time?

Weld-On actually melts the peices together removing the crack all together. But even if it does extent it will be on the dry side of the O-ring groove so it shouldn't be a problem. I have the King Pin running right now Folding. I can't do any tests for now though. My thermo probes are all messed up. Each one is about 3C off of each other and they are not rising and falling at the same rates so they can't be calibrated. :rolleyes: I am working on it but it looks like one of those "make the wallet much lighter" deals which isn't a good thing right now with Christmas almost here. I got a few dead mobos that I am going to unsolder the thermisters from and try the joy stick mod fom bittech. If I deam that unusable then at least I will have learned something and it will not have cost me anything. :D

I got an interesting project on the drawing table to make use of some of that lexan you sent me. Will be working on that come Turkey day weekend. :cool: The GPU block came out awsome aswell. Still need to finish the top but the base is sweet. Messed up the NB block though. I will go over that later though. Time to go to bed now. Long weekend...

CPU, GPU and NB block. :) NB block is going oto be re-done though. I messed it up.
http://www.customcomp.us/kingpinpic/006.jpg

Great work jaydee. Those blocks look great!

Anyway, why does the NB block need to be redone? I cant see a visible mistake.

[first post!]

Perhaps he needs to leave more room for an O-ring groove next time.:shrug:

2 things. One) I didn't get the pattern square in the base so I have to shave off a little and scrap the O-ring and would have used silicone instead, but that brings me to Two) The mounting holes are backwards. :cry: I decided to re-do it right after that but...

I just remilled a sweet looking one with O-ring groove and all but when I was tapping the hole I slipped, broke the tap, and sliced my finger to the bone right on top of the nuckle with the jagged end of the broken tap. So now I got a sweet NB block with a broken tap that is to small to be drilled out. :mad: So that is #2 in the scrap pile. I did however complete the GPU block and it looks good. In fact it is so overkill it should cool a CPU better than the CPU block I made. :rolleyes:

Bah. Well I might do another NB block if I get around to it but I am running out of time. I only got a few more hours of mill time up here and I think that will be spent finishing King Pin #2. I don't even need a NB block for that system anyway being it isn't going to be OC'ed much. I might do one anyway.... King Pin #2 look good. If I get a good scanner pic I will start a new thread tonight. If not I will start it when I get home Sunday. I think #Rotor will like it.

Jaydee,

I did that knuckle slicing on my right hands "bird" a couple years ago. Not good, it was slow to heal. Hope you had it stitched up as that helps to limit it breaking open.

Blocks all looking good, glad to hear the GPU & CPU performance are where you want them to be.

Here is a couple sneak peaks at a couple blocks I finished this weekend.

And another. I did one with the pins and I did one with just channels. I will get more spacific later in a new thread. I got to pack up and go home now. I will try to get the info up in the next fews days on these.

:eek: You've got to be kidding me JayDee...that is exactly, EXACTLY the block I was designing on paper over here during my down time. I bet the dimensions of the internals are somewhat different but the concept is EXACTLY THE SAME!!!! God damn thats crazy! I was thinking of drilling the holes around the pins with a 1/8" end mill then connecting the holes to form the channels with a small, perhaps 1/16" or 1mm bit. Overall dimensions of the pins array would turn out to be 1.2" x .9". I know between jack and squat about machining Cu though so I am probably smoking crack on that, but still, that is exactly it. I was thing the barb reliefs on either side of the channels to be .9" x .6" then have base thickness in the barb reliefs and channels to be app. 2mm and overall height of the Cu piece as 6mm. Then have 1/4" NPT tapped holes in the Plexi top with a matching .9" x .6" x .15" relief over the matching reliefs in the base plate. I am rambling and probably don't make any sense....damn stumped again, so much for that design...... :rolleyes:

Yeah, I used 1/8" drill bit for the main holes and a 1/16" end mill to connect the holes. The pinned one is 3/8" and drilled down 1/4" for a 1/8" base thickness. The other one with just channels wasw the same process but i used 1/4" material and drilled 3/16" and left 1/16" for the base. I will test it out and then next time I get on the mill I will finish connecting the holes to make pins and retest it to see if it helps or not. Also have a center inlet jet deal i am thinking up.

That is one thing about this hobby, the block you have in mind has probably already been done. :D This block is a mixed concept block, it uses #Rotor design, White Water design, RBX design, and even some of my own thinking to boot. :D

Man Jaydee, you turn out new variations of blocks fast. Used a ball tip end mill on the pin block, or at least that the way it looks.

Did you also use the ball tip mill on the block with the irreguler channels?

I hope you have time to run some of these blocks on your test bench soon.

Nice work!