I am trying to build a system that uses a flow meter pulse to control voltage up and down.

Envisioned
Lipo battery connected to a "unit" to adjust voltage up or down
the "unit" is controlled by the pulses coming from the slow meter
the "unit" is read by a voltmeter.

a lipo battery 11.7v plugged into some sort of "unit" that adjusts the output voltage from 100% (11.5v) down to say about 50% or (5.5v)
the pulses coming from the flow meter controls the unit. lets say a fluid flows 1 way for X revolutions the voltage will go up to 100% or 11.1v if the flow is reversed the voltage will drop down to 50%

Is there something that already does this some sort of counter that controls voltage output? if there is what is this called, if not is there anyone out there who can help me build this.

 

At first glance, a design to do what you require does not seem too demanding. The one stall point might be to devise some method to signal when the flow rotor reversed direction. For the initial, forward flow pattern, the pulse signal would simply serve as the clock to a counter chip, which would activate a flip-flop when a pre-set number is attained.
So, does this brief description look like it meets your requirements?

 

At first glance, a design to do what you require does not seem too demanding. The one stall point might be to devise some method to signal when the flow rotor reversed direction. For the initial, forward flow pattern, the pulse signal would simply serve as the clock to a counter chip, which would activate a flip-flop when a pre-set number is attained.
So, does this brief description look like it meets your requirements?

Hello DCR can you explain more how to du this? Electronics are not my strongest point. Basicly in a mechanical system I was thinking of a flow meter controlling a potentiometer to increase or decrease voltage. I need to do the same on a digital system

 

One approach is to have pulses from the flow meter increment a binary counter, and the binary output drives a D/A converter, and that analog voltage is offset and scaled appropriately for the output. Sounds like a lot, but it might be pretty simple. First, there are a zillion counters to choose from, and many of them will run on 12 V directly. Next is the D/A. If you don't need very high resolution, like only 16 steps from the min to max output voltages, then the D/A can be a few resistors. Next is the output section, probably one opamp or a dual opamp.

1. What is the resolution you need in the output analog voltage? What is the voltage change for each flow meter output pulse?
2. Does the max output voltage have to be so close to the power rail voltage? 0.2 V is not much headroom for whatever the output circuit is.

ak

 

One approach is to have pulses from the flow meter increment a binary counter, and the binary output drives a D/A converter, and that analog voltage is offset and scaled appropriately for the output. Sounds like a lot, but it might be pretty simple. First, there are a zillion counters to choose from, and many of them will run on 12 V directly. Next is the D/A. If you don't need very high resolution, like only 16 steps from the min to max output voltages, then the D/A can be a few resistors. Next is the output section, probably one opamp or a dual opamp.

1. What is the resolution you need in the output analog voltage? What is the voltage change for each flow meter output pulse?
2. Does the max output voltage have to be so close to the power rail voltage? 0.2 V is not much headroom for whatever the output circuit is.

ak

I want to use this flow meter http://www.conrad.com/ce/en/product...-PP-BIO-TECH-eK-FCH-M-Duese-1-mm-0015-08-lmin

 

I want to use this flow meter http://www.conrad.com/ce/en/product...-PP-BIO-TECH-eK-FCH-M-Duese-1-mm-0015-08-lmin

Does the flow meter detect direction? I only see one output, power and ground.
So two flow meters would be required and a directional valve each way?
A counter and a DAC (Digital to Analog converter) and a voltmeter.

The flow meter would detect the number of times the liquid flowed which way but not the volume.

 

One approach is to have pulses from the flow meter increment a binary counter, and the binary output drives a D/A converter, and that analog voltage is offset and scaled appropriately for the output. Sounds like a lot, but it might be pretty simple. First, there are a zillion counters to choose from, and many of them will run on 12 V directly. Next is the D/A. If you don't need very high resolution, like only 16 steps from the min to max output voltages, then the D/A can be a few resistors. Next is the output section, probably one opamp or a dual opamp.

1. What is the resolution you need in the output analog voltage? What is the voltage change for each flow meter output pulse?
2. Does the max output voltage have to be so close to the power rail voltage? 0.2 V is not much headroom for whatever the output circuit is.

ak

A

Does the flow meter detect direction? I only see one output, power and ground.
So two flow meters would be required and a directional valve each way?
A counter and a DAC (Digital to Analog converter) and a voltmeter.

It has to be done with only 1 flow meter, the sensor inside the flow meter is a hall effect switch. based on my expierence (limited as it is) hall effects can detect both ways I think, any time a magnet is passed by it. there are 10,000 pulses per liter. I do not know what the impulse voltage is at the moment.

 

Hey Guys,
How do I program the binary clock and the digital analog converter? I basicly need it to step voltage 8 times, but could use a 16 step
How do I lay the circuit do I need anything else like diods.....

 

A


It has to be done with only 1 flow meter, the sensor inside the flow meter is a hall effect switch. based on my expierence (limited as it is) hall effects can detect both ways I think, any time a magnet is passed by it. there are 10,000 pulses per liter. I do not know what the impulse voltage is at the moment.

It can detect both ways but it can not tell one direction from another.

 

At first glance, a design to do what you require does not seem too demanding. The one stall point might be to devise some method to signal when the flow rotor reversed direction. For the initial, forward flow pattern, the pulse signal would simply serve as the clock to a counter chip, which would activate a flip-flop when a pre-set number is attained.
So, does this brief description look like it meets your requirements?

At first glance, a design to do what you require does not seem too demanding. The one stall point might be to devise some method to signal when the flow rotor reversed direction. For the initial, forward flow pattern, the pulse signal would simply serve as the clock to a counter chip, which would activate a flip-flop when a pre-set number is attained.
So, does this brief description look like it meets your requirements?

DRC

So what you are saying, is that even though the rotation might be reversed, and if the flow meter does not have reverse signal the, but is still picking up off the hall effect, that at 100% it would flip the counter/ voltage down is this correct? can this be programmed with a 100% and say 10% flips?

 

Hey Guys,
How do I program the binary clock and the digital analog converter? I basicly need it to step voltage 8 times, but could use a 16 step
How do I lay the circuit do I need anything else like diods.....

CD40192 or CD40193 counters. Nothing to program, unless you want an Arduino, or some such programmable design. CD40192 is a decimal counter (0 to 9). CD40193 is hexadecimal (0 to 15).

If it must be done with one flow meter you just need to monitor which way the pump is going to control the up or down count. That can be done depending on the pump.

 

Use an LM3914 led driver chip after the DAC device. Then you have the option of either having a visual led for each voltage increment or else a current pulse to activate whatever. The 3914 chip can be set up for 8 or more voltage increments. It uses an analog input, and has several adjustable components.
Any chance of adding some sort of internal directional sensor to the flow path of the tubing? A pressure sensor placed at each end of the flow meter might be able to pick up a pressure differential which would change with flow direction. ... Or maybe an orifice with a differential pressure sensor.
http://www.ti.com/lit/gpn/lm3914

 

A


It has to be done with only 1 flow meter, the sensor inside the flow meter is a hall effect switch. based on my expierence (limited as it is) hall effects can detect both ways I think, any time a magnet is passed by it. there are 10,000 pulses per liter. I do not know what the impulse voltage is at the moment.

So do you need a count that goes up to 10,000?
If you have the counts pulses from the pump why do you need the flow meter?

 

Here is a differential pressure sensor, just for example. It would require tapping into the flow path, at two points, to get the differential pressure. It appears to be capable of 1 psi resolution. ... still looking around, though.

http://www.te.com/usa-en/plp/ms4515/Xoye8v.html

... So, is the fluid you are dealing with corrosive, toxic, or something else to be concerned about?

 

DRC

So what you are saying, is that even though the rotation might be reversed, and if the flow meter does not have reverse signal the, but is still picking up off the hall effect, that at 100% it would flip the counter/ voltage down is this correct? can this be programmed with a 100% and say 10% flips?

Why don't you just describe your goal and the layout of the pumping equipment and flow meters.

Right now it seems like you have a half-baked idea of what is possible and we seem to be trouble-shooting a half-baked solution rather than designing something with you to fill all of your needs.

1) why do you have a pulse from 5 to 12v instead of a digital pulse that is on/off
2) what other signals are possible that can tell a system to reverse or forward pump.
3)

A bunch of AND or OR gates or a single microcontroller can be programmed to make almost anything - 40 pins of glorious digital or analog inputs and outputs and hundreds of lines of code that can make that pump dance or sing if you want it to. Again, describe your problem, not your solution.

 

If he/she fancies the idea to be patentable they might be reluctant to put out the whole idea. but are there such things already on the market?

 

So do you need a count that goes up to 10,000?
If you have the counts pulses from the pump why do you need the flow meter?

There is no pump the liquid is syphon fed from .01 psi up to 8 psi depending on how far a valve is opened

 

There is no pump the liquid is syphon fed from .01 psi up to 8 psi depending on how far a valve is opened

That's nice. Are you going to answer any of the other questions in posts 2, 3, 4, 11, 13, and 15?

ak

 

Hey guys
Sorry it was very late last night (my time) when I posted.
Here is some more details.

1. The unit must be very small for robotics it is on a DC system, power supply 11.5 thru with a step down chip to 5v to operate the flow valve signal.
2.There is no pump the liquid in the resivour is pressureized by air from .01 to 8 psi depending how far a valve is opened
3. it must be able to measure .3oz per minute up to 8oz per minute
4. the liquids are flammable, 1 is petroleum based, the other is alcohol based
5. there can only be 1 flow meter, due to space restrictions.
6. The flow meter in the post has a pulse signal of 10000 pulses per liter, it uses hall effect sensors to create the pulses
7 It must be able to measure from 0 oz to between 2 and 32 oz and everything in between depending on the size of the resivour.
8. it needs to be adjustable I.e. if a 2oz tank is used 100% voltage out put would be shown when full and 0 voltage output when tank is empty, if a larger tank is used it must be easily adjustable to show 0%(5v)empty and 100% (11.1v) full.
9. if a tank is partially filled say 90% it must show 90% voltage on the voltometer, and it must be able to count down backwards from that 90% till it hits 0%
10. the out put voltage to the voltometer can a continuous drop or be stepped between 8 and 16 times, (the voltage is dropped 16 times by .4v increments starting at 11.7v-11.3 -10.9-10.5......
For continuous; lets say you have a 2oz resivour the minimum the flow sensor can read is .3oz per minute. at minimum flow it would take 6 minutes to flow out. at maximum flow it would take 5 seconds. That said the sensor does 1 rotation pulse for every .00338oz (10,000 rotations per litre, 1 litre = 33.8oz / 10,000 = .0038oz/pulse) passed thru. The unit is sensitive for 100 rotations or .33oz So if you have 2 oz that means that a full tank will have 526 rotations/pulses. Now the voltage out to the voltometer can be dropped from 11.7v 100% output down to 5v output 0% with over 526 pulses which would be .0126v per rotation. Or It can be stepped down by 16 times 526/16= per 33 pulses the voltage is dropped down by .418v

Because the volume can change issue is it has to be adjustable, lets say I choose to use a 48oz tank. this would mean that it would have a total of 48oz / .00338oz/pulse = 12631 pulses for a 48oz tank 11.7v-5v= 6.7v /12631 pulses = .00053044097v/pulse OR the out put voltage could be dropped 16 times = 713 pulses voltage is dropped by .418
11. Thus It the unit must be able to count the number of pulses going into the resivor and then subtract that controlling the output voltage going to the voltometer automaticly.
12. There could be a switch to flip from the counters + count to the negative count

 

There is no pump the liquid is syphon fed from .01 psi up to 8 psi depending on how far a valve is opened

Then is it impossible to get a directional signal from the siphon?