Definitely love the idea.... currently using the mCubed flowmeters and they themselves cripple flow. An alternative that'd work with the same hardware would be invaluable...

It's simple. The SF800 produces a 100Hz signal per litre per minute flow. So interpreted as an RPM value by the mobo, and by MBM, it produces 6000 RPM per litre flow. In MBM you can divide the RPM value by 6. You then get, effectively, ml per minute flow. You just display this in Samurize using the 1000 decimal dot. Presto.

http://img.photobucket.com/albums/v1.../25_resize.jpg

http://img.photobucket.com/albums/v1.../51_resize.jpg

http://img.photobucket.com/albums/v1.../07_resize.jpg

This is a Proteus Industries 200 Series flow meter . The spinning impeller contains magnets at the tips, which generate a current in the coil that is located in the housing of the meter, separated from the chamber. The current, generated in the coil by the Hall effect, is directly proportional to the rotational speed of the impeller, thus being linearly indicative of the flow through the meter. The second half of the flow meter contains all the electronics that interpret the current and convert it to a 0-5VDC analog output. It also contains a relay with an adjustable trip point.

I connected the Red Lion DC voltmeter to read the analog output from the flow meter. Also, this model of the meter can be set to scale the readout, and this is how I calibrated the meter to display 100 under normal circamstances. The buzzer is connected externally and is activated when the relay inside the flow meter senses flow that is below a set point.

The flow meter is able to output analog voltage signal that is proportional (very linearly) to the flow, and I have calibrated the panel meter to show 100, as in 100 percent, under normal conditions. I can adjust the voltage supplied to the pump through a pot, and the meter mirrors the reduced flow quite nicely, bottoming out at 36%, at which point the relay inside the flow meter trips off another relay that is connected to a loud buzzer, giving a nice warning signal.

[quote=Rezistor]

This is a Proteus Industries 200 Series flow meter
Looks nice, but they want $250+ for it... Not exactly low cost.

[quote=jaydee]Ebay hooked me up with this one. I do agree wholeheartedly about the MSRP, the only explanation I can come up with that could start to explain the excessive price is the original intent for industrial application, along with the installation, callibration, maintenance, and tech support. But honestly, building the circuit board inside probably won't cost more than $20, and milling out a piece of 1/2" thick plastic should be even cheaper.

[quote=Rezistor]Yeah, it is an interesting idea. I would probably use delrin or even brass for the housing. I like the idea of seeing the flow aswell.

[quote=jaydee]I've got Proteus 100 Series flow meter with brass housing, and I would not recommend a brass housing for computer use. It weighs 2 pounds! It needs some serious support, you can't just let it hang on the hose, it'll put immense stress on the barb that's attached to that piece of hose.

[quote=Rezistor]2 pounds of brass shouldn't be needed for this application. :shrug: IMO the meter should be as small as possible to fit in all cases not just monster cases. It should be no larger than 2" x 2" x 1.5" thick and preferably smaller. Shouldn't weigh more than 1/2 pound. Delrin would probably be the thing to use. Anything larger than a MCP350 pump would be to big and take up to much room in mid to small cases. If they want to sell a lot of these then they need to make them fit in as many variety of cases possible. Seeing the flow isn't necessary either but would be nice.

And as for that link to wizd forums... ( http://www.wizdforums.co.uk/showthre...ight=flowmeter )

What he's trying to do has already been done and is commercially available already at low cost... and has been for... er... about 4 years I think (as long as the Innovatek Rev3 waterblock hsa been in existence) - See http://www.webshop-innovatek.de/0000...3e0eac52f.html

They also do this one - http://www.webshop-innovatek.de/0000...3b0dc5e17.html
and this one - http://www.webshop-innovatek.de/0000...3e0e2802b.html

All in the region of 20 euros... ish... and all can be gotten in the US from http://www.highspeedpc.com/

Jaydee,

Don't know if you gave up on the idea of buying a Swissflow flowmeter or not, but Elfa (www.elfa.se), a European distributor sells it.
As for the shipping costs, it's the same story as when i want to buy something from the U.S.
http://www.elfa.se/images/lowres/l2153.jpg

Ok Marci, no problem, I was just trying to help out.

If there is already a solution out there, then not much use in making a mold.

Unless you think using jewel bearings will be much of an improvement?

Since I do not have any of these meters, I do not know if there high quality.
Our expensive lab meter has jeweled bearings, and very un-restrictive flow, and I am basically copying that design using magnets we are already producing.

That's a nice site. I can't buy anything from anywhere but the USA though. I have to have detailed and accurate delivery time on shipping as I travel for work and must be home when it arrives or it will be shipped back as there is no one here to receive it. I only get about a week a month were I know I will be home.

Dont stop mate, those innovatek ones kill flow big time, Ive tried them before & lost 50% of flow because of them..

low loss flowmeter would be a biggie for me & many others Im sure.

Yep, I was referring to the guy on WizD forums' project - basically a waste of time when it's already been done, however the Innovatek ones are 1/4" BSP for 8mm OD Tubing, so imagine the ID if you will ;)

A cheap 1/2" ID suitable alternative is certainly required... and I'd definitely have a load off you Dave...

However, rest of those in the uk, if desparate, head to rswww.com and type "compact flow sensor" into the search box.

Retails at £41 for one off, £39 for 6 off. Certainly an alternative for those in the UK...

They also do one suitable for -25 deg C thru 125 deg C, pressure drop of 1bar at 1cSt - £78 a piece.

They do a dirt cheap one for 15mm diam tubing, if you can adapt it down it may be suitable... only £20.37. Type "Liquid Flow Sensors -15mm dia Pipe" into their search box.

Basically, rswww.com > Automation > Fluid Management & Control > Flow Transducers (if you want to walk yer way thru the catalog and see all the options)

looks good marci, cheers for that.

Just ordered one, will hook up to my fanomatic pro & see how I get on (controller allows you to enter the pulses/litre of the flow sensor so should be reasonably accurate)

Aye, have used the fanomatics in the past... let us know how you get on...

Guys at DD seem to want a flow meter, so I will go ahead with a mold.

There is no big developement costs here, I have been doing this on my own, using parts AVT already developed, and C-Systems has 1000's.

I designed for a cheap two side MUD insert, so no one here will mind if I use one, and even if we only sell 100 of them, that will cover on mold cost :)

Just out of curiosity? How much will it cost?
How low cost is it

^ For a mold blank insert? about $500.
Plastic molds are cheap, if there simple two peice inserts.

It is when you get into 3 or more peice molds, does it get very expensive.
It is all about design and CNC time, which I am doing for free.

We use MUD systems on all our injection machines
http://www.masterunitdie.com/

I ment the final "Low Cost Flow Meter"
how much is low cost...

If it is around $50 or less and works decent I will buy at least 2.

I was hoping for the neighborhood of $20 - doesn't need to be uber accurate.

As far as the guy on wizd's project - the boards hes making also work on the paddle sensors that have larger (3/8") barbs. The circuit (which is an optical sensor, not like the innovatek products which are AFAIK mechanical) can be used on any paddle wheel based meter.

if it is $20 or less I will buy 5.

Ditto. that's about how many I'd need.... well... 1 for 1/2" and then 4 for 6mm ID.

If DD is going to b the reseller then there is no way they will be under $40 IMO though. Maybe we can get them directy from Dave for less?

Read the page I linked you to - Innovatek's aren't mechanical, they're IR, and their "available separately" control board used to convert a flow indicator into a flowmeter can also be added to any paddle wheel.

One nice thing about paddle wheel flowmeters is that they're like a good old fashioned 'visual flow indicator' : walk into a room, take a quick or lazy glance that a-way, and feel reassured. Ones with clear plasic or glass on both sides of the rotor let you mount an LED behind them so you can do this in the dark.

If you can take the face off to remove a stray string of teflon tape or whatnot without having to pull it from the loop, that can be convenient too.

Plastics like nylon that can soak up surficants, antioxidants, anticorrosives, whatnot beyond my chemistry / materials testing are a bummer.

Is there a reliable spec. for the bottom end of pulses / sec. that an hypothetical generic fan header should be expected to sense? ( I have a gems rota-flow installed on one old MB that can't sense as slow as the flowmeter signals; it's a fanles / silent rig that could get by on 0.5 lpm. )

Tangent: I'd write off designing a flowmeter around producing an off the shelf correlation between 4-20 mA pulses / sec and lpm or ml/s or gpm of a particular coolant mix of a particular viscosity with a particular thermal conductivity at a particular temperature and pressure. Like 'low cost flowmeter' on the one hand and the lab kit and skills to establish all that on the other...

I suppose it would be possible to design one that worked in only one flow direction, or was good at trapping bubbles in any but one precise vertical alignment -- that would be sad.

That's what comes to mind this afternoon for someone trying to design / build / sell a low-cost paddle wheel meter for this niche.

Price / performance wise the Swissflow does look interesting, and if it puts out a ball park accurate flow based on a common fan divider, well that is nice indeed.

I'm hoping that being in Canada that i can get them directly from Dave and save myself the customs charges...

It does. It is very accurate, connects straight to a mobo fan header and to get ml/min. you just divide the signal by 6 in MBM. I've been messing about with Samurize a bit more:

http://pic18.picturetrail.com/VOL932.../101207747.jpg
(Temps are a bit high today: 24C ambient room temperature...)

Ideally I'd like a nicely engineered paddle flow meter with a window, LED lighting, accurate IR sensing with a straight conversion of RPM to litres/min flow (or GPM, depending on the flick of a switch or jumper), and integrated temperature sensor. Even better if the thing doubles as an airtrap/reservoir. Even better still if you could bolt it straight onto a pump like the Laing DCC or D5.

We can but dream... but since I'm planning a new project and I know someone with amazing CNC skills, I think I may start designing something myself...

Digiflow 8000T
 

I just recieved one ($59+shipping). It looks to be an excellent piece. Hooked it up to an Eheim 1048, resevour and radiator and it read 0.48gpm. It has a totalizer and a timer. It appears to be rather restrictive. It is about 3/8" at the narrowest where the impeller is located.