Overclock.net › Forums › Cooling › Water Cooling › DO NOT use XSPC res tops with your Swiftech pumps...
New Posts  All Forums:Forum Nav:

DO NOT use XSPC res tops with your Swiftech pumps... - Page 4

post #31 of 80
[quote=Digidoc;11992831]Well it depends.

Quote:
If you slap a supercharger on your engine and you throw a rod, then yeah, you're going to have to fight the dealership. If you change the tires though and your engine all of a sudden develops a problem with the piston rings, the dealership and MFG are going to have a very hard time denying your warranty, especially if you remind them of the Magnuson-Moss Warranty Act and what the number is of your local district attorney.
This would depend on what size of tire you have installed on your new car
remember if your car came with 14 inch rims and you put 22's on it you are
causing a drastic change in effective gear ratio thus affecting engine load
Pa's Puter
(15 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 i930@4.2ghz. Gigabyte EX58-UD4P EVGA GTX 970 FTW + GSkill DDR3-1600 12 gig 
Hard DriveOptical DriveCoolingOS
Samsung Evo SSD  DVDROM EK supreme HF Cu gold plated/ XSPC 480 Windows 7 Pro. 64 bit 
MonitorKeyboardPowerCase
LG 27EA63 IPS monitor merc stealth Corsair TX850 Corsair 700D 
MouseMouse PadAudio
logitech G5 Rocketfish dual surface Creative Xfx extreme and Zalman 5.1 headphones 
  hide details  
Reply
Pa's Puter
(15 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 i930@4.2ghz. Gigabyte EX58-UD4P EVGA GTX 970 FTW + GSkill DDR3-1600 12 gig 
Hard DriveOptical DriveCoolingOS
Samsung Evo SSD  DVDROM EK supreme HF Cu gold plated/ XSPC 480 Windows 7 Pro. 64 bit 
MonitorKeyboardPowerCase
LG 27EA63 IPS monitor merc stealth Corsair TX850 Corsair 700D 
MouseMouse PadAudio
logitech G5 Rocketfish dual surface Creative Xfx extreme and Zalman 5.1 headphones 
  hide details  
Reply
post #32 of 80
Quote:
Originally Posted by Digidoc View Post
...I'm failing to see how the pump top could be the cause of it. If anything, replacing the top with one that has less restriction should have the opposite effect that Swiftech is claiming. The pump would now not have to work as hard to draw in water from the reservoir, meaning the pump should run cooler.

...Additionally, there's no directly link between the impeller and the electronics that drive the impeller (they're magnetically linked)....

...Given that the pump's magnetic drive runs at a constant speed and is physically disconnected from the rest of the pump, if the rotor was out-spinning the magnetic field that's driving it (highly unlikely), it wouldn't be able to "push" the magnetic drive. If it did, it would most likely induce a back-current back into the magnetic coil. If you had a multi-pump setup I could possibly see that happening, but in a single pump setup it's not going to happen.
A pump "volute" or "impeller well", as its called, can easily cause pump failure, even if it is less restrictive. I design custom volutes for larger motors and pumps to change their flow-pressure bias curves, and there are limits. Usually, I have to rework the impeller as well, and paying attention to the electrical characteristics is most important. The work I do is mostly on AC permanent magnet motors and on DC with "smarter" motor controllers (the same thing if you know motors), but from what I have seen, these are simplified permanent magnet motors with a DC chopper circuit and voltage oscillator. There doesn't appear to be any feedback compensation or anything of the sort (it would be obvious), so these pumps are running the same sort of circuit that controls your case fans. Its going to run hot no matter what, but with such low voltages its usually not a concern.

This is bad news for those like me who are used to tinkering with the 'wet-end' of the pump. Without changing the electronics, or at least being able to hook up some probes to a scope to see what the electrical characteristics are like for the stator nodes, tampering with the flow characteristics is an easy way to blow the motor.

Im not sure here, as the terms being used here do not indicate the true nature of the volute design. "More flow" and "less restrictive" aren't really terms that have meaning with this science since they are relative to each other. For example, if I designed the pump volute to have more pressure bias in a highly restrictive loop, then I would actually see more flow... so then its "high flow" too, right? The flow and pressure of the pump depend on the entire closed loop for any real conclusion. If the pump's volute is enlarged for more flow bias, it could lose its pressure bias, and then on a more restrictive loop it would actually pump less than the stock volute. That is not all though... if the volute is enlarged, and the impeller is the same, it also becomes easier for the impeller to spin faster despite the relative restriction. In this case, the smaller volute that allows less water to flow around the impeller in the pump would generate more pressure and also flow, and the pump with a larger volute would perform worse. The case you imagine with two pump in series is very similar and with an oversized volute on a highly restrictive loop its actually very possible for the impeller to 'slip' ahead of its usual position with respect to the current in the coils. I would have to see a side by side of the modded volute and the stock one to draw any conclusions though.

I can tell you though that running a volute that is LESS "restrictive" is actually an easy way to blow this sort of motor. Its not a true DC motor with brushes. In those, as load is removed, the speed does increase. With these, the rules for synchronous AC motors are applied (because thats what they really are despite the term "DC"). All this relating to automotive COMBUSTION engines should be put to rest. These are very different, and the logic that I see applied here (from both sides) so far is incorrect. There may not be a "physical" connection between the stator and rotor, but neither is there between the spark plugs, fuel, and air to the crankshaft, nor is one needed to cause physical damage (or then running jet fuel in a car engine wouldn't be a problem, would it?). Both motors are examples of "energy conversion devices" but their respective rules are different. Magnets and inductors ARE the linkage, and as such, their EM forces represent the transfer of energy... by the very logic that there is force present to rotate the motor rotor with some 'invisible force', it is not logical to then state that there is no way for damage to result since there is no mechanical linkage. Such a conclusion denies the very mechanism that allows the motor to work in the first place.

For this, the magnet can 'overspin' or 'slip' if its load is too light, and if the pump is designed for a given load at a given speed, if that load is too light, it will reach its radial axis points too fast for the electrical cycle, causing it to become a generator from that point on until the phase from the electronics flips. This will cause "back EMF" to be generated into the coils. This kills the motor's efficiency for one, but it also means energy is being put back into the stators, which results in more heat being generated right in the stators. This repeated 'stutter' can cause cyclical failure from the heat buildup, melting the stator coils from the inside out. Considering water cooling in general, I would not seek a pump volute that provides more flow unless I knew its pressure generation were kept the same. I would look for a pump 'top' that provides the same head pressure w/ respect to back-pressure, and then any additional flow claims can be claimed as innovative. But I suspect that in this case, the greater flow may have been achieved by lowering the pressure handling characteristics, which when run on a more restrictive loop means LESS performance and the ability for the impeller to slip ahead more... resulting in a 'meltdown'.
post #33 of 80
Please let me comment on the following then to clear up some of the bad info and get you guys on the right path about motors. I dont mean to pick on anyone... for the most part, I think digidoc is really close, but some of his conclusions are just a little off:

"A magnetic drive motor spinning at full RPM without a load will not heat up the electronics that drive the stator. If anything, it'll have the opposite effect: it should run cooler since there's no load holding back anything." -Digidoc

Okay, think of the motor from the thermodynamics POV (first law of thermo) and you will realize how wrong this is. Feel free to face-palm because Im sure you will realize how you insulted your own intelligence by ignoring this basic law. If the motor has power (watts) being put into it, and it usually does say... 20 watts of work on the load and 10 watts is lost to self-inductance... then has the 20 watts of load removed... where will all that heat go exactly? Depending on the motor type, it may consume a different wattage or speed up (but not with a synchronous motor like here), but its not going to be a cooler even in a relative sense. The self inductance will rise (wattage may drop or stay same but VA will rise) and internal heat buildup will rocket. Try running a similar aquarium pump out of water and watch it melt its coils, if taking it out of water doesn't cause it to sputter and vibrate so bad it meets its end from cyclical mechanical wear. Due to the input of this motor being a DC based chopper circuit (most likely) with no feedback correction, the effects of too much slip and heat buildup are amplified.

"When you replace the top on the 35x series, you're not actually modifying the pump so the impeller spins any faster than before." -Digi

Maybe not in rpm (steady-state) since a permanent magnet means its a synchronous motor, but the speed in terms of rotor position within each cycle... yes.. you are changing this. In general, its actually considered more damaging to underload a pump/motor than it is to overload it. Just something for you to mull over.
Edited by undertheradar - 1/13/11 at 6:17pm
post #34 of 80
I would like to see that the manager of XSPC would like to say about this , As i know him , i shall call him and see, XPSC make quality components and to be hones if you modify your pump by using an XSPC Res top , you void your warranty any ways so if you ask me its a fools game fooled once wont be fooled again, but im positive XSPC make quality components and i would like to hear their response on this before people jump to conclusions.
Black Noir CosmoS
(13 items)
 
  
CPUMotherboardGraphicsRAM
i5-750 Asus P7P55D PRO 5850 CrossXFire 8GB Gskill Ripjaw 
Hard DriveOptical DriveOSMonitor
Gskill Pheonix 40GB SSD Pioneer Win 7 64Bit Sony 19" Viao Sony SDM-E96D 
KeyboardPowerCaseMouse
G15 V2 750 OCZ Faita1ty Custom Cosmos S Roccat Kone V1 
Mouse Pad
Roccat Taito 
  hide details  
Reply
Black Noir CosmoS
(13 items)
 
  
CPUMotherboardGraphicsRAM
i5-750 Asus P7P55D PRO 5850 CrossXFire 8GB Gskill Ripjaw 
Hard DriveOptical DriveOSMonitor
Gskill Pheonix 40GB SSD Pioneer Win 7 64Bit Sony 19" Viao Sony SDM-E96D 
KeyboardPowerCaseMouse
G15 V2 750 OCZ Faita1ty Custom Cosmos S Roccat Kone V1 
Mouse Pad
Roccat Taito 
  hide details  
Reply
post #35 of 80
OK so i have been following this thread since the very beginning. I have come to the conclusion that some people in here have absolutely no idea what they are talking about and some that know what they are talking about.

From one perspective, the pump can spin at any speed and the electronics won't know, and it won't effect it in anyway.
My counter to that is, with an electric motor, the magnets and the electro magnets are not connected, so the spinning magnetic field should always spin at the same speed, but yet, if you torque down an electric motor with a lot of stress, what happens? it gets hot.

so in turn if you put more stress on the DDC pumps, they will get hotter.

then you look at this, if you decrease resistance through the entire pump top, you are allowing more water to go into the pump, which takes more power to move, so in turn, will stress the pump more.
so the added water flow from the top is causing the pump to become stressed more, and heat up, which then, possibly kills it.
post #36 of 80
After reading your posts Undertheradar I would first like to applaud you for going to the effort to write all of that down!

Secondly, in answer to question on Pump tops and performance here is a pressure/flow graph for the MCP 355 + A bunch of tops:



As you can see the pump tops improve the maximum head pressure and maximum flow of the pump quite a bit. By reading your post it seems to me that as long as you are under 1.5 GPM with the stock top you should be fine, but at higher rates of flow the pump will start to degrade. Please correct me if I'm wrong.
Labor of Love
(14 items)
 
  
CPUMotherboardGraphicsRAM
2600k @ 5.0 GHz/1.42v ASUS P8P67 Pro MSI GTX 670 OC 20 GB/1600/7-7-7-24 
Hard DriveHard DriveOptical DriveOS
120GB Vertex SSD Western Digital 1TB None! Ubuntu 12.04/Win8/OSX 10.8.2 
MonitorKeyboardPowerCase
27" 1440p X-Star OCN Mechanical - MX Browns Seasonic X-560 800D w/RX480 top mounted 
Mouse
Logitech USB Generic 
  hide details  
Reply
Labor of Love
(14 items)
 
  
CPUMotherboardGraphicsRAM
2600k @ 5.0 GHz/1.42v ASUS P8P67 Pro MSI GTX 670 OC 20 GB/1600/7-7-7-24 
Hard DriveHard DriveOptical DriveOS
120GB Vertex SSD Western Digital 1TB None! Ubuntu 12.04/Win8/OSX 10.8.2 
MonitorKeyboardPowerCase
27" 1440p X-Star OCN Mechanical - MX Browns Seasonic X-560 800D w/RX480 top mounted 
Mouse
Logitech USB Generic 
  hide details  
Reply
post #37 of 80
Quote:
Originally Posted by GingerJohn View Post
I'm a bit late to the game I know, but it is actually the other way round. The pump will use MORE power at higher flow rates (lower restrictions), not less. Counter intuitive I know, but this has been proven by Martin's test. Note that this graph shows the results of under- and over-volting the pump and is used here to illustrate that the power draw increases with increasing flow rate:



If you go down to the table at the bottom of Martin's test you can get the full test report for the XSPC Res Top and the MCP 355 stock top.

From the individual test results (links above) you can compare the power draw at max flow. The pump with the XSPC top pulls ~28W, compared to the stock ~20.5W.

With the XSPC top you will be going above the stock maximum of 20.5W at ~2GPM, so if the pump is running above that then maybe it would lead to the increased rate of failures.

Although if the rate of failures has recently increased that would point to a change in manufacture or components.

I would be very interested to see what Swiftech say.
Those test results are inconclusive as far as the electrical parameters are concerned. Since its DC going in, we have no idea what the actual voltage/current phases are with respect to rotor position. The input to the pump itself, after the controller, is what would tell us more. The power going in might be increasing because the 'VA' of the pump, not the actual wattage. This is the problem with perm-magnet pumps that have the electronics built into the housing... we need to monitor the connections between the control electronics and the rotors. I have seen perm-magnet motors shoot up in wattage when the load is removed all together. These motors are really still AC motors (sinusoidal or whatever waveform), just with a DC input to the controller, but since the VA must be handled by the controller just like the wattage, a motor with the load removed all together can slip so far out of phase that its peak voltages increase even though its RMS drops, and since this all needs to be rectified at the controller, you cant draw any conclusions about this sort of motor/pump without more data than just the input power to the controls.
post #38 of 80
Quote:
Originally Posted by SilveR_172 View Post
I would like to see that the manager of XSPC would like to say about this , As i know him , i shall call him and see, XPSC make quality components and to be hones if you modify your pump by using an XSPC Res top , you void your warranty any ways so if you ask me its a fools game fooled once wont be fooled again, but im positive XSPC make quality components and i would like to hear their response on this before people jump to conclusions.
I would ask if he actually had an engineer sign off on the mod or if this was done like so many 'backyard engineer' mods. Just because it works in one situation doesn't mean it will work in everyone else's. The modded pumps need to be tested with a wide range of restriction just like you would see IRL. I suspect this 'high flow' top might limit the range of back-pressures that this pump can safely operate with. While it is possible to improve on the pressure and flow characteristics of some OEM stuff by making the internals more slick & rounded or making the bends longer for less losses, what you are more likely doing if you see any improvement is making the pump more suitable for a more narrow range of operating conditions. The OEM design may not be optimal for YOUR setup because they have to cover a wide range of possible conditions. Im sure I could mod one of these pumps to give 500gph of flow at the same wattage! But it would require 1" pipe and as soon as you put a restrictive waterblock or radiator on it, the impeller would slip so far ahead (more distance between the impeller blades and impeller well) it would melt down in this application. More 'flow bias' can really work against you if you try to hook it up to a system that is more restrictive than usual.
Edited by undertheradar - 1/13/11 at 6:45pm
post #39 of 80
Quote:
Originally Posted by charliehorse55 View Post
After reading your posts Undertheradar I would first like to applaud you for going to the effort to write all of that down!

Secondly, in answer to question on Pump tops and performance here is a pressure/flow graph for the MCP 355 + A bunch of tops:



As you can see the pump tops improve the maximum head pressure and maximum flow of the pump quite a bit. By reading your post it seems to me that as long as you are under 1.5 GPM with the stock top you should be fine, but at higher rates of flow the pump will start to degrade. Please correct me if I'm wrong.
I know this complicates things, but I would have to see the current/wattage of the pump to see how it changes as the pressure/flow changes. If the pressure and flow both increase and the input wattage is constant, then it is a good thing. I doubt these pump's controls have current regulation built in though...
post #40 of 80
I actually don't think it is XSPC's fault at all.

The oem restrictive top adds resistance to the spinning motor, increasing current draw, which in turn increases heat through the coils of the armature, commutator and brushes of the electric motor in the pump.

By using a top with lower restriction, you are in turn reducing the resistance placed on the motor and that reduces current draw, which reduces heat.

Reducing heat from the core components of the electric motor in the pump should actually improve the electric motor's longevity, not decrease it

Therefore if damage is occurring prematurely, even though there is now less heat being drawn and longevity should improve, it is from either:
1. incorrect reassembly
2. pump circuitry manufacturing fault
3. pump motor manufacturing fault
4. poor quality or faulty part used in pump manufacture
5. overrated pump cycle life
6. overheating due to environment
7. coolant or moisture getting into circuitry of pump


P.S. Almost all failures of the Swiftech pumps have been show to be the FETs on the circuit board - so I'll blame poor circuit design and cheap shoddy circuit components before I'd blame any third party top
Crimson Basilisk
(25 items)
 
  
CPUMotherboardGraphicsGraphics
Intel i7 2600K @ 5.1Ghz AsRock Extreme4 Gen3 Palit GTX 680 Jetstream 2GB Palit GTX 680 Jetstream 2GB 
RAMRAMHard DriveHard Drive
G.Skill F3-12800CL9D-8GBXL G.Skill F3-12800CL9D-8GBXL Sandisk Extreme 120GB SSD Sandisk Extreme 120GB SSD 
Hard DriveOptical DriveCoolingCooling
WD Caviar "Black" 1TB SATAII LG WH14NS40 Blu-ray Burner XSPC Rasa CPU (Acetal) Black RX360 Triple Fan Radiator 
CoolingCoolingOSMonitor
XSPC DDC Two BayRes BMaverick Laing DDC (MCP350) Windows 7 Ultimate x64 Philips 201B 21" CRT 
MonitorKeyboardPowerCase
ViewSonic VX922 19" LCD Diamond Digital Black Antec True Power 750w Antec LanBoy Air Red 
MouseMouse PadAudioOther
Epic Gear Meduza (By GEIL) Steel Series QCK+ XL Asus Xonar DX ROCCAT Kave 5.1 Surround Sound Gaming Headset 
Other
Sennheiser hd555 
  hide details  
Reply
Crimson Basilisk
(25 items)
 
  
CPUMotherboardGraphicsGraphics
Intel i7 2600K @ 5.1Ghz AsRock Extreme4 Gen3 Palit GTX 680 Jetstream 2GB Palit GTX 680 Jetstream 2GB 
RAMRAMHard DriveHard Drive
G.Skill F3-12800CL9D-8GBXL G.Skill F3-12800CL9D-8GBXL Sandisk Extreme 120GB SSD Sandisk Extreme 120GB SSD 
Hard DriveOptical DriveCoolingCooling
WD Caviar "Black" 1TB SATAII LG WH14NS40 Blu-ray Burner XSPC Rasa CPU (Acetal) Black RX360 Triple Fan Radiator 
CoolingCoolingOSMonitor
XSPC DDC Two BayRes BMaverick Laing DDC (MCP350) Windows 7 Ultimate x64 Philips 201B 21" CRT 
MonitorKeyboardPowerCase
ViewSonic VX922 19" LCD Diamond Digital Black Antec True Power 750w Antec LanBoy Air Red 
MouseMouse PadAudioOther
Epic Gear Meduza (By GEIL) Steel Series QCK+ XL Asus Xonar DX ROCCAT Kave 5.1 Surround Sound Gaming Headset 
Other
Sennheiser hd555 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › DO NOT use XSPC res tops with your Swiftech pumps...