need a 1/2" flowmeter - Page 2

MadDogMe · 11-12-2002, 09:22 AM

Quote:

MDM
why not a couple of relays in sereis with a solid state time delay between them ?

I can't get my head round it I'm afraid, I've been awake too long. wherever I look I see a open relay that prevents the PC being turned on, we need to stop it activating the sensor cutout so there is time to turn it on without switching straight off, BUT there's no power to even start with to acheive that!...

The switch in the sensor would have to be reversed(so that flow loss would close the switch). so that the PC power relay was closed/activated by default, and flow stopping would make the sensor switch connection and 'power' it open, long enough for the PC to turn off. it would need to be in a 'power off' after powercut state as well to prevent it starting straight up again. would that work?.

Or would it work if we used the power wire on the PSU instead of cutting/relaying the AC/mains?...

PS, The more I think about the push switch I doubt I'd find a suitable 240AC one, one that you could run parallel with DC current

...

PPS. I n---e----e---d S---------l----ee-----e---p :0 zzzzZZZZZ

gone_fishin · 11-12-2002, 04:38 PM

Here is an interesting pdf file pertaining to the subject.

click me

BillA · 11-12-2002, 05:47 PM

a sales document
"Adherence to installation guidelines provides the best
possible installation and the closest correlation to published values. Precise K-Factors are best derived from on-site
calibration using a calibrated weight or volume measure versus time."

and thats as close to accuracy that they get
I still hold that +/-1% is 'hype' for any paddle wheel

gone_fishin · 11-12-2002, 05:57 PM

I tend to agree with you. I found the document informative on the actual functioning of the flowmeter since I have no first hand knowledge of one of these.

bigben2k · 11-12-2002, 06:08 PM

Turbine meters seem to be most accurate, if built correctly. Not too terribly restrictive, but still there.

nOv1c3 · 12-04-2002, 08:08 PM

Came across this while browsing and thought of this thread ..A little diaphragm-operated pressure switch at $1:50

hook it up to a relay to kick your power off ...If your pump stops or you lose pressure in your system ..

Dont know if it would work but at a $1:50 its somthing to play with lol

Oops help if i post the link

http://www.allelectronics.com/cgi-bi...-10&type=store

BladeRunner · 12-05-2002, 05:28 AM

That could work and make a very cheap protectection system used with a pulse gen, and mbm5 with shutdown, (both free apps). could be done for under $5 total I guess

It's a shame the site selling can't be arsed to find out the specs though

be nice to know the reaction preasure from definatley on / definately off

Since87 · 12-05-2002, 09:06 AM

Quote:

Originally posted by nexxo
Several threads exist on this subject. One referred to Dwyer as the place to go and it is indeed recommended!

This is excellent. An EE I used to work with, now works for them. Maybe I can get a scratch and dent item or two.

Since87 · 12-05-2002, 09:15 AM

As far as protection is concerned, I think a thermal switch mounted to the waterblock, and wired to power down the computer in the event of overheating is the safest and simplest way to go.

The fundamental purpose of the protection is to protect the processor from overheating.

Pump fails - thermal switch saves the CPU.

Rad gets totally clogged with dust - thermal switch saves the CPU.

Apartment next to yours is on fire and ambient gets to 80C - thermal switch saves the CPU. (At least temporarily.)

Thermal switches are available in a large range of trip temperatures at Digi-Key.

bigben2k · 12-05-2002, 10:00 AM

I e-mailed Fairchild about the pressure switch.

I like the thermal switch, I just wish it didn't have to be wired into the PSU, but that is by far the safest, as long as the switch is mounted correctly.

I'd still rather try to emulate Intel's throttling scheme. Maybe when I get my mobo...

bigben2k · 12-05-2002, 10:59 AM

Got some news.

Fairchild says that it's not a Fairchild part, it's actually made by "MCE/KDI Resister Products".

I turned up nothing on their 6 websites.

BladeRunner · 12-05-2002, 11:16 AM

The thermal switch could be substituted for the flow switch, or even added to my solution switching the "pulse gen" so enabling a soft shutdown. The flow switch is just that, a simple and reliable switch type operated by coolant flow, any form of switching could be added or used exclusively. I've also got a liquid switch that I'm going to put in the res, but I'll probably just connect this as a separate alarm function only using the system fan header and a pulse gen.

I guess in most systems the temp switch on the block would be just fine. I prefer a primary solution that starts to take action as soon as the flow has stopped for whatever reason, rather than waiting for a secondary, changeable reaction like overheating. This is even more important if High power TEC's are to be used, and to be honest an instant, "pull the power plug to everything" is the best way for that, considering a High watt TEC can start to unsolder a waterblock fairly quickly if the coolant flow stops!.

Since87 · 12-05-2002, 02:13 PM

Quote:

Originally posted by BladeRunner
The thermal switch could be substituted for the flow switch, or even added to my solution switching the "pulse gen" so enabling a soft shutdown.

It would be easy to add a watchdog timer chip like this.

This chip monitors a pulsing signal. If the pulses stop (for a length of time determined by a capacitor value) the watchdog output pin of the part changes state.

The watchdog output could control a relay powering the computer. If the computer doesn't manage to shut itself down in lets say 10 seconds. The watchdog chip goes ahead and cuts power anyway.

I don't know about you, but I wouldn't want to count on anything involving Microsoft when it comes to the life of my CPU.

bigben2k · 12-05-2002, 02:28 PM

Halfway there... now to patch the watchdog into the system clock!

BladeRunner · 12-06-2002, 07:13 AM

What you say is inherently true but highly unlikely. Firstly Aquarium pumps are by their very design extremely reliable items, (if you are using a quality type like an Eheim). Secondly Windows by design does like to lock up when shutting down occasionally, however I'd put money on the Pump being the more reliable of the two, The odds of both crapping out together must be extremely high, I'll take that chance.

The point is I have my protection up and running, it's working at this moment, and would power down my PC now if the pump failed or flow stopped. Some of the ideas posted here are brilliant, but ideas wont power down the PC, and maybe we are bordering the line of dismissed returns with all this.

If someone can show me a proven cheaper simple way that is as reliable or more so, I'll congratulate them and use myself, until then I'm happy to trust my windows app based system in the very unlikely event the pump should die, because it's far better than nothing at all , I test it at least twice a week by stopping the flow. So far it's never failed to shut off the system in the proper way within 20 seconds

bigben2k · 12-06-2002, 09:39 AM

That's all good BladeRunner, and I'll probably incorporate that, in one form or another, but I still want to give the slow system clock a shot.

Does anyone have any info on the chip that drives the system clock?

BladeRunner · 12-06-2002, 10:35 AM

bigben2k

I was more replying to Since87, than you, and I'm interested in what you'll find with system clock device, it's just getting a little to "electronic" for my understanding now

11-12-2002, 06:08 PM	#30
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Turbine meters seem to be most accurate, if built correctly. Not too terribly restrictive, but still there. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

12-04-2002, 08:08 PM	#31
nOv1c3 Crazy Stupid Join Date: Nov 2002 Location: Dallas texas Posts: 149	Came across this while browsing and thought of this thread ..A little diaphragm-operated pressure switch at $1:50 hook it up to a relay to kick your power off ...If your pump stops or you lose pressure in your system .. Dont know if it would work but at a $1:50 its somthing to play with lol Oops help if i post the link http://www.allelectronics.com/cgi-bi...-10&type=store Last edited by nOv1c3; 12-04-2002 at 09:44 PM.

12-05-2002, 05:28 AM	#32
BladeRunner Cooling Savant Join Date: Jul 2001 Location: Chesterfield Uk Posts: 459	That could work and make a very cheap protectection system used with a pulse gen, and mbm5 with shutdown, (both free apps). could be done for under $5 total I guess It's a shame the site selling can't be arsed to find out the specs though be nice to know the reaction preasure from definatley on / definately off __________________ Zero Fan Zone

12-05-2002, 10:00 AM	#35
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	I e-mailed Fairchild about the pressure switch. I like the thermal switch, I just wish it didn't have to be wired into the PSU, but that is by far the safest, as long as the switch is mounted correctly. I'd still rather try to emulate Intel's throttling scheme. Maybe when I get my mobo... __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

12-05-2002, 10:59 AM	#36
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Got some news. Fairchild says that it's not a Fairchild part, it's actually made by "MCE/KDI Resister Products". I turned up nothing on their 6 websites. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

11-12-2002, 04:38 PM	#27
gone_fishin Cooling Savant Join Date: May 2002 Location: Da UP Posts: 517	Here is an interesting pdf file pertaining to the subject. click me

11-12-2002, 05:47 PM	#28
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	a sales document "Adherence to installation guidelines provides the best possible installation and the closest correlation to published values. Precise K-Factors are best derived from on-site calibration using a calibrated weight or volume measure versus time." and thats as close to accuracy that they get I still hold that +/-1% is 'hype' for any paddle wheel

11-12-2002, 05:57 PM	#29
gone_fishin Cooling Savant Join Date: May 2002 Location: Da UP Posts: 517	I tend to agree with you. I found the document informative on the actual functioning of the flowmeter since I have no first hand knowledge of one of these.

12-05-2002, 09:15 AM	#34
Since87 Pro/Guru - Uber Mod Join Date: Sep 2002 Location: Indiana Posts: 834	As far as protection is concerned, I think a thermal switch mounted to the waterblock, and wired to power down the computer in the event of overheating is the safest and simplest way to go. The fundamental purpose of the protection is to protect the processor from overheating. Pump fails - thermal switch saves the CPU. Rad gets totally clogged with dust - thermal switch saves the CPU. Apartment next to yours is on fire and ambient gets to 80C - thermal switch saves the CPU. (At least temporarily.) Thermal switches are available in a large range of trip temperatures at Digi-Key.

12-05-2002, 11:16 AM	#37
BladeRunner Cooling Savant Join Date: Jul 2001 Location: Chesterfield Uk Posts: 459	The thermal switch could be substituted for the flow switch, or even added to my solution switching the "pulse gen" so enabling a soft shutdown. The flow switch is just that, a simple and reliable switch type operated by coolant flow, any form of switching could be added or used exclusively. I've also got a liquid switch that I'm going to put in the res, but I'll probably just connect this as a separate alarm function only using the system fan header and a pulse gen. I guess in most systems the temp switch on the block would be just fine. I prefer a primary solution that starts to take action as soon as the flow has stopped for whatever reason, rather than waiting for a secondary, changeable reaction like overheating. This is even more important if High power TEC's are to be used, and to be honest an instant, "pull the power plug to everything" is the best way for that, considering a High watt TEC can start to unsolder a waterblock fairly quickly if the coolant flow stops!. __________________ Zero Fan Zone

12-05-2002, 02:28 PM	#39
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Halfway there... now to patch the watchdog into the system clock! __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

12-06-2002, 07:13 AM	#40
BladeRunner Cooling Savant Join Date: Jul 2001 Location: Chesterfield Uk Posts: 459	What you say is inherently true but highly unlikely. Firstly Aquarium pumps are by their very design extremely reliable items, (if you are using a quality type like an Eheim). Secondly Windows by design does like to lock up when shutting down occasionally, however I'd put money on the Pump being the more reliable of the two, The odds of both crapping out together must be extremely high, I'll take that chance. The point is I have my protection up and running, it's working at this moment, and would power down my PC now if the pump failed or flow stopped. Some of the ideas posted here are brilliant, but ideas wont power down the PC, and maybe we are bordering the line of dismissed returns with all this. If someone can show me a proven cheaper simple way that is as reliable or more so, I'll congratulate them and use myself, until then I'm happy to trust my windows app based system in the very unlikely event the pump should die, because it's far better than nothing at all , I test it at least twice a week by stopping the flow. So far it's never failed to shut off the system in the proper way within 20 seconds __________________ Zero Fan Zone

12-06-2002, 09:39 AM	#41
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	That's all good BladeRunner, and I'll probably incorporate that, in one form or another, but I still want to give the slow system clock a shot. Does anyone have any info on the chip that drives the system clock? __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

12-06-2002, 10:35 AM	#42
BladeRunner Cooling Savant Join Date: Jul 2001 Location: Chesterfield Uk Posts: 459	bigben2k I was more replying to Since87, than you, and I'm interested in what you'll find with system clock device, it's just getting a little to "electronic" for my understanding now __________________ Zero Fan Zone

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