chiller & pump progress

bigben2k · 09-20-2003, 09:13 PM

I'm still "toying" with my chiller, and while I was trying to figure out how to make a pump cram flow through what is probably a 1/4" diameter chilling area, it hit me that instead of using valves to set the flow rate, I could simply vary the voltage to a DC pump.

I'll report results as I progress.

joemac · 09-20-2003, 09:20 PM

Have you check to see if you can set the flow rate via a menu through the 232 port? A lot of times all that is required to communicate with these types of devices is hyper terminal.

bigben2k · 09-21-2003, 10:27 AM

Hum... could be worth a shot. I don't expect that the rate is variable, but I'll check (hopefully today) and report.

bigben2k · 09-24-2003, 10:45 PM

The chiller works.

I had to open it, to put it in a loop, because I didn't have the matching quick-disconnects. (Some pretty gruesome pics to come later!)

I filled it up with distilled water, which contained a bit of Jet-Dry, and fired it up.

Going through the controls, I set the temp to 10 deg C (it defaults at 30 deg C), and noticed that there was a provision for a "gradient", of up to 60 deg C per hour.

Then I sat back and watched the "Manifold" temp drop. It first drops by 0.07 deg C, then by 0.06 three times, and then found that it would repeat the same pattern. I quickly checked the water temp "finger style", and everything seemed to correspond (FWIW).

This chiller can also heat the water (but who cares?!?). That part works too. There's also an air suction and exhaust connection, which I'm told is part of a "dryer" section, but was faulty. I have no information on it, other than that it was added to the basic model for a custom built unit for GM (General Motors). Fittings are 3/8 quick disconnects (male) and fit in the water connections, but the air hoses are 1/4" OD, so the fittings are useless.

My next step is to get an accurate thermometer, and then try to correlate the temp readings with actual values. I'll then try to swap out the pump, open up the tubing to 3/4" where possible, and see if the impact is going to prevent me from using the chiller that way. I already contacted the manufacturer, and have enough info to proceed.

All internal (water) tubing is 3/8" ID, except for the pump inlet which appears to be 1/2" ID. All tubing is insulated with 1/4" thick neoprene (single piece) and joints are carefully taped with 1/8" sticky neoprene. Tie wraps everywhere... Nice.

The original pump is a Gorman Rupp Industries (GRI) model# 17651-142, 12 Vdc (? amps). From the website, it appears to be a magnetically driven centrifugal pump. The bottom of this pump actually rotates openly. I will try to get the specifications to this particular model, if they are available (probably not). Specs for the whole chiller include "avg flow: 0.5 gpm". My plan is to swap it out for the infamous Johnson pump, and control the flow by varying the voltage (there's no sense in adding heat from extra pumping action, when it's not needed, and can be avoided).

The cooling unit is made up of two very large heatsinks, sandwiched together and bolted. It's apparent that the gap was then filled with and expanding liquid insulating foam, for a nice tight seal (I am not taking that apart!). The cooling unit also has two 4" Papst fans, running side by side, at 12 volts DC and drawing ~0.3 amps each.

I have yet to try to hook up to the RS-232 interface (where's that darn adapter?!?), but I don't hold much hope for it.

That is all.

joemac · 09-24-2003, 10:50 PM

Wow – I can’t wait to see what that unit looks like on the inside. I am amazed at my ability to resist opening it up. Can’t wait for the pic’s

, any new updates ?

bigben2k · 11-08-2003, 01:20 PM

Ok, here it is, finally!

The two heatsinks sandwich the pelts. There are 2 * 4" fans on one side of it.

murray13 · 11-08-2003, 07:39 PM

Ben, here is some info about the T-252P you might like to know.

They were made for General Electric Medical Systems.

There are 10 (yes ten) 50W pelts wired in series inside of the chiller assy. At max cooling they have about 65V going to them. Yes they run them at about half power, which is close to best efficiancy.

Just a note, that power supply for the pelts has exposed 120VAC and 65VDC on the board so be careful when it's out of the case.

The water input and output of the chiller assy are 1/4NPT. Pretty small. I have modded one to use 3/8NPT but it wasn't easy. There's not a lot of clearance making the holes bigger. What we will do to keep flow resistance to a minimum...

The temp sensor used is spec'd at +/- 1deg C, but over the temp range used it is +/- .5 deg C.

Running a stock T-252P and an original Swiftech WB cooling just a 2100+ @ stock speed and V with folding running as a service and a room temp of 21C the display says the coolant is at 10.83C.

You may know some or all of this, but if there is anything else you would like to know, I just might have allready looked into it. I do have three of them. One running and two more for parts/moding.

I did have one of the chillers all the way apart after running it for over 2 1/2 years and found that Thermotek only uses silicone to seal the nylon barbs into the block. Had a very small leak inside all of that foam. Got to the pelts and corroded one to the point of failure.

Oh well... that's what mod'n is all about right? Making things either work better or look better.

joemac · 11-08-2003, 09:31 PM

High voltage + water

Ben take a pic of that power supply - 10 pelts - nice

bigben2k · 11-09-2003, 10:15 PM

Thanks for all the tips!

I thought they were made for Chrysler, but couldn't imagine what they would do with them, especially the air suction bit. Any ideas about that? I'm about to pick up a vacuum gauge, to see what it can do.

The first thing I'm going to do, is do a temporary (aka non-permanent) conversion to 3/4", to adapt the Johnson pump. If I succeed, I'll make the conversion permanent, and go with it.

Were you able to get past the 5 deg C increments, in settings the temps? I'm thinking about hacking the BIOS.

bigben2k · 12-03-2003, 11:02 AM

A small update:

The control panel only has 4 buttons:
-pump start/stop
-temp up
-temp down
-output temp

The thing is, the last button does more than one thing. It can switch the unit to "Gradient mode" or "Stabilized mode", but I came across these quite accidentally, by repeatedly pressing that button. I haven't been able to reproduce the sequence yet.

On a positive note, the "temp up/down" buttons can actually allow me to set the temp in 0.5 deg C increments. Once again, it appears to be in the timing of how I press the buttons.

I did notice that the display occasionally flickers at a 1 Hz rate, so that's going to be my starting point, in figuring out this unit's functions.

More as I progress.

09-20-2003, 09:13 PM	#1
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	chiller & pump progress I'm still "toying" with my chiller, and while I was trying to figure out how to make a pump cram flow through what is probably a 1/4" diameter chilling area, it hit me that instead of using valves to set the flow rate, I could simply vary the voltage to a DC pump. I'll report results as I progress. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

09-21-2003, 10:27 AM	#3
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Hum... could be worth a shot. I don't expect that the rate is variable, but I'll check (hopefully today) and report. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

09-24-2003, 10:45 PM	#4
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	The chiller works. I had to open it, to put it in a loop, because I didn't have the matching quick-disconnects. (Some pretty gruesome pics to come later!) I filled it up with distilled water, which contained a bit of Jet-Dry, and fired it up. Going through the controls, I set the temp to 10 deg C (it defaults at 30 deg C), and noticed that there was a provision for a "gradient", of up to 60 deg C per hour. Then I sat back and watched the "Manifold" temp drop. It first drops by 0.07 deg C, then by 0.06 three times, and then found that it would repeat the same pattern. I quickly checked the water temp "finger style", and everything seemed to correspond (FWIW). This chiller can also heat the water (but who cares?!?). That part works too. There's also an air suction and exhaust connection, which I'm told is part of a "dryer" section, but was faulty. I have no information on it, other than that it was added to the basic model for a custom built unit for GM (General Motors). Fittings are 3/8 quick disconnects (male) and fit in the water connections, but the air hoses are 1/4" OD, so the fittings are useless. My next step is to get an accurate thermometer, and then try to correlate the temp readings with actual values. I'll then try to swap out the pump, open up the tubing to 3/4" where possible, and see if the impact is going to prevent me from using the chiller that way. I already contacted the manufacturer, and have enough info to proceed. All internal (water) tubing is 3/8" ID, except for the pump inlet which appears to be 1/2" ID. All tubing is insulated with 1/4" thick neoprene (single piece) and joints are carefully taped with 1/8" sticky neoprene. Tie wraps everywhere... Nice. The original pump is a Gorman Rupp Industries (GRI) model# 17651-142, 12 Vdc (? amps). From the website, it appears to be a magnetically driven centrifugal pump. The bottom of this pump actually rotates openly. I will try to get the specifications to this particular model, if they are available (probably not). Specs for the whole chiller include "avg flow: 0.5 gpm". My plan is to swap it out for the infamous Johnson pump, and control the flow by varying the voltage (there's no sense in adding heat from extra pumping action, when it's not needed, and can be avoided). The cooling unit is made up of two very large heatsinks, sandwiched together and bolted. It's apparent that the gap was then filled with and expanding liquid insulating foam, for a nice tight seal (I am not taking that apart!). The cooling unit also has two 4" Papst fans, running side by side, at 12 volts DC and drawing ~0.3 amps each. I have yet to try to hook up to the RS-232 interface (where's that darn adapter?!?), but I don't hold much hope for it. That is all. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

09-24-2003, 10:50 PM	#5
joemac Cooling Savant Join Date: Dec 2002 Location: Dallas Posts: 339	Wow – I can’t wait to see what that unit looks like on the inside. I am amazed at my ability to resist opening it up. Can’t wait for the pic’s , any new updates ? __________________ www.aquajoe.com

11-08-2003, 01:20 PM	#6
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Ok, here it is, finally! The two heatsinks sandwich the pelts. There are 2 * 4" fans on one side of it. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

09-20-2003, 09:20 PM	#2
joemac Cooling Savant Join Date: Dec 2002 Location: Dallas Posts: 339	Have you check to see if you can set the flow rate via a menu through the 232 port? A lot of times all that is required to communicate with these types of devices is hyper terminal.

11-08-2003, 07:39 PM	#7
murray13 Cooling Savant Join Date: Sep 2002 Location: Omaha, NE USA Posts: 216	Ben, here is some info about the T-252P you might like to know. They were made for General Electric Medical Systems. There are 10 (yes ten) 50W pelts wired in series inside of the chiller assy. At max cooling they have about 65V going to them. Yes they run them at about half power, which is close to best efficiancy. Just a note, that power supply for the pelts has exposed 120VAC and 65VDC on the board so be careful when it's out of the case. The water input and output of the chiller assy are 1/4NPT. Pretty small. I have modded one to use 3/8NPT but it wasn't easy. There's not a lot of clearance making the holes bigger. What we will do to keep flow resistance to a minimum... The temp sensor used is spec'd at +/- 1deg C, but over the temp range used it is +/- .5 deg C. Running a stock T-252P and an original Swiftech WB cooling just a 2100+ @ stock speed and V with folding running as a service and a room temp of 21C the display says the coolant is at 10.83C. You may know some or all of this, but if there is anything else you would like to know, I just might have allready looked into it. I do have three of them. One running and two more for parts/moding. I did have one of the chillers all the way apart after running it for over 2 1/2 years and found that Thermotek only uses silicone to seal the nylon barbs into the block. Had a very small leak inside all of that foam. Got to the pelts and corroded one to the point of failure. Oh well... that's what mod'n is all about right? Making things either work better or look better.

11-08-2003, 09:31 PM	#8
joemac Cooling Savant Join Date: Dec 2002 Location: Dallas Posts: 339	High voltage + water Ben take a pic of that power supply - 10 pelts - nice __________________ www.aquajoe.com

11-09-2003, 10:15 PM	#9
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Thanks for all the tips! I thought they were made for Chrysler, but couldn't imagine what they would do with them, especially the air suction bit. Any ideas about that? I'm about to pick up a vacuum gauge, to see what it can do. The first thing I'm going to do, is do a temporary (aka non-permanent) conversion to 3/4", to adapt the Johnson pump. If I succeed, I'll make the conversion permanent, and go with it. Were you able to get past the 5 deg C increments, in settings the temps? I'm thinking about hacking the BIOS. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

12-03-2003, 11:02 AM	#10
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	A small update: The control panel only has 4 buttons: -pump start/stop -temp up -temp down -output temp The thing is, the last button does more than one thing. It can switch the unit to "Gradient mode" or "Stabilized mode", but I came across these quite accidentally, by repeatedly pressing that button. I haven't been able to reproduce the sequence yet. On a positive note, the "temp up/down" buttons can actually allow me to set the temp in 0.5 deg C increments. Once again, it appears to be in the timing of how I press the buttons. I did notice that the display occasionally flickers at a 1 Hz rate, so that's going to be my starting point, in figuring out this unit's functions. More as I progress. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)