Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

help needed with 12 volt fluid flow sensors and pulse generators

Status
Not open for further replies.

garyd

Member
Hi , a couple of months back I posted a question regarding capactive fluid level sensors ( they turned out to be conductive) and received the help I needed in fooling an external PC water cooling unit into thinking it had the level sensors connected when it actually did not , the system has worked fine now without any alarms which would occur if the sensors were not connected, I now have one further modification which I would like to do in that the system uses a fluid flow meter of the rotating vane type which also has an LED this connects to the units circuit board and then to an IC which controls and monitors pumps fans flow and level sensors and temperature sensors I want to ditch the flow meter as I have another within the computer itself and I do not need 2 which is detrimental to good fluid flow within the cooling loop and having heavily modded the external cooler there is now no space for it's flow meter and i've had to mount it outside which is far from ideal , my limited electronics knowledge is basic at best but I have determined that it may be possible to replace the flow meter with a pulse generator and have purchased one via ebay it has 3 pins VCC GND and OUT but I am unsure as to how I would connect it to the external coolers flow meter plug (socket) the flow meter has 4 wires red black white and yellow which obviously carry the supply current the LED's power and also the pulse signal from the flow meter, Multimeter readings from the 4 wires are as follows Red = 4.88 volts Black = GND White = -018 volts and Yellow = constantly fluctuating between 1.88 and 2.05 volts I do not need the LED function and want to eliminate that part entirely but am unsure as to whether it is the Yellow or the White wire which supplies it's power and also whether it is the White or Yellow wire which I should connect along with the Black and Red to the pulse generators 3 pins ie VCC GND and OUT .Any help or thoughts would be very much appreciated , Gary .
 
Last edited:
Yoru sentences are very run-on which makes what your trying to say garbled.

You did not link to the pulse generator to answer such things as what is the range of Vcc.

Now, I might expect the value of your sensor to be closer to 5.00 rather than 4.88. It's a little more than 10% low.

Not knowing what you have, I'm hesitant of giving you this information. If you can light a LED with your new found pulse generator, you could be "in like flint".
A LED requires a series resistor and abut 10 mA usually. Now, I'm lost - no information)

Most sensors are either a transistor (commonly called open collector) that closes to ground or a switch that closes to ground. Once you get a LED to pulse, you can replace the LED with an OPTOcoupler such as a 4N26 or something better. The Emitter of the optocoupler would go to ground and the collector would be the output that simulates the sensor.

SO my suggestions here use a separate power supply for the function generator (my crystal ball is broken (MCMisB)). I'm assuming it can source 10 mA ( (MCMisB) etc. etc. Concerns are, we don't have enough info to go on. IF you can operate the function generator on the 4.88 V, then the opto can be elimiated and replaced by a resistor and an NPN transistor like the 2n3904. You could have used a 555 timer or the low power CMOS version, the 7555 timer.

I was hoping to find a schematic for a sensor. e.g. **broken link removed** here it says 5-24V and current sourcing and sinking, The sinking is the NPN transistor.
 
Last edited:
Hi , thanks for your reply , I have again measured the voltage and it's steady at 4.88 I only mentioned that the flow meter has an LED because I thought it might help in determining which wires go where so to speak , as for my English I write they way I was taught with punctuation marks rather than using seperate lines , the link to the pulse generator is as follows **broken link removed** and I repeat that I do not intend to use an LED I just needed to know whether it was the White or the Yellow wire which supplied it with current .
 
the 0P said:
Multimeter readings from the 4 wires are as follows Red = 4.88 volts Black = GND White = -018 volts and Yellow = constantly fluctuating between 1.88 and 2.05

Can you clarify the -18 Volts? White could be common.

Can you measure relative to white and relative to black?

1) functioning normally
2) Flow stopped
3) Sensor unplugged
 
the readings I gave are the ones obtained with the unit functioning normally however I have now noticed that if I put my multi meters probes onto the boards contacts with the Black Probe contacting the GND pin and the Red probe contacting the White pin I get a reading of 0.18 volts but there is a minus sign to the left of the display which normally indicates reversed polarity if however I connect the Red probe to GND and the Black probe to the White pin it still displays 0.18 volts but the minus sign has now gone ?. The readings for the unit with pumps stopped and the alarm sounding are GND to White =1.0 to 4.95 fluctuating this is with the probes going Black to White pin and Red to GND pin and again it gives a minus sign if I connect the probes normally the readings are the same with the flow meter disconnected however I have noticed that if I take a reading across the GND and RED pins whilst the unit is alarming and the flow meter is disconnected it initiates the reset and the alarm stops and stays stopped until I remove the probes .
 
Gary, you had said 18 in post #1 with a leading 0, "White = -018 volts" That's why I asked. I suspected 0.018.

repeating what you said, tabulated:

Red = 4.88 volts
Black = GND
White = -.018 volts
Yellow = constantly fluctuating between 1.88 and 2.05 volts

What I would EXPECT is from black to yellow with the sensor unplugged is something relatively steady. 4.88 or closer to 0.6V. Afterall, what would be causing it to "pulse"?
Now, they could use the White as a "sensor connected" signal.

One other thing you can try is to power the bought pulse generator from GND and 4.88 V. It should operate that low. See if the LED on the generator pulses at ow frequency or remains steady at high frequencies. Monitor the 4.88 supply so see if it dips appreciably.
 
Black to Yellow with flowmeter unplugged and alarm sounding it is fluctuating wildly between zero and 4.95 Volts but if I press the reset button it then cancels the alarm for a few seconds and the voltage is rock steady at 4.95 Volts ,I have connected the pulse generator with Red to VCC Black to GND and Yellow to OUT and the LED pulses but only at about 1 pulse per second which I believe would represnt too low a flow and sound the alarm which it does , I have tried moving the jumper from it's as supplied position which is closest to the edge of the PCB one poition further in towards the centre of the PCB one place at a time until I reached the last position but things only seemed to get worse , I have not touched either of the 2 variable Resistors as to be honest I have'nt a clue as to whether to adjust one or both and by how much ? , I am also wondering if the fact that during these tests the White wire being connected to nothing is in itself causing the unit to alarm ?.
 
Note that you have jumpers that select the range.

The potentioneters are likely 10 turns ad will start to "click" at their stops.

The pot labeled CW duty cycle decrease, means when you turn it left, the on time of the LED will decrease. e.g. from 1 sec on to 1 sec off to 0.5 s on to 1.5 sec off.

Similarly the "CW frequency increase" will increase the frequency when turned clockwise from 1 to 50 Hz, if on that range.

The jumpers will just make the LED on steady and probably decrease the brightness, because your eye can't see the changes.


==

You have a long time where you can edit posts. Usually, I delete the contents and mark them "<duplicate>" and ask the mods to delete one.
(I just asked for removal of the dup)
 

Attachments

  • 14008963.jpg
    14008963.jpg
    35.5 KB · Views: 149
Last edited:
Ok i'll have a play around with it and see what adjusting the cw frequency does but it still does'nt help me in knowing what to do with the White wire ?.
 
I have a Maplin electronics store quite close to me and althouhgh i'm disabled due to a Stroke I should be able to get someone to go there for me , will I need anything else such as a circuit board and if so what type ?.
 
I'm afraid my son is away at the moment and can't help me with the bits but I can order them on ebay it'll just take a couple of extra days , I need to know though the Wattage of the Resistors you want me to get, is 1/4 Watt ok ?
 
1/4 to 1/2 W should be fine. Type not important either: metal film, metal oxide, carbon film, carbon composition are all OK. +-5% is normal these days. Just need one resistor value between 330 and 420 ohms. You probably want a package with leads and not an surface mount package. You might also want to pick up a simple red LED as a troubleshooting aid.

I have no idea about the white wire either. Remember, we are guessing and you can't measure stuff with a scope.
 
ok i'll get some 1/4 watt metal film ones as they seem the most common on ebay and i'll let you know when I have them and the Transistor which will probably be by the coming weekend .
 
Can you post a link to the flow sensor?
 
I should perhaps have added that Koolance sell a frequency adaptor board that allows you to plug the Flow meter into a MOBO 4 pin PWM fan header and read the flow as if it were a fan , this was not needed with the external cooler as apart from the plug it worked with the exteral coolers electronics straight out of the box .
 
The manual for that Koolance model is here. It is clear the the yellow and white wires have no sensor-produced voltage on them and are simply for the two contacts of a normally-open reed switch. The red and black are respectively the + and - connections for the LED supply (which you presumably will be using instead to power the pulse generator; red to Vcc and black to ground).
So here's how you could try interfacing the pulse generator output to the cooler input :-
FlowSensorInterface.gif
Note that the white wire is grounded here. Both yellow and white need to connect to the cooler input.
Q1 acts like the reed switch (which the manual actually refers to also as 'open collector').
Depending on the cooler input circuitry, connecting the pulse generator 'Out' terminal directly to the cooler input, as you did, may have made it unhappy, because the generator is based on a 555 circuit which actively pulls 'Out' high.
 
Last edited:
Ah that's interesting i'm not familiar with reed switches very much but guess that the magnet on the Impellor opens and closes it as it passes , i'm not very good with electronic schematics so will try and brush up on it a bit in order to fully understand the connections I will need to make .
 
The manual states "The Koolance INS-FM16 uses a magnetic reed
contact switch (open collector)". The options are very different, but generally compatible. I picked a smaller resistor from (4.88-0.6)/10e-3).

You could measure the resistace to the controller (sensor output pins) to +5 (sensor disconnected) for use to get a better idea. It could tell us the pull up resistor value. One value should be pretty steady and probably around 1K. The output pin, would be the one that was varying.

Alec's circuit was what I was eluding too. The 1/6 74xx figure I posted too.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top