How Can I Design a Digitally Controllable Constant Current Source?

[jim hardy]

Great job on that schematic.

I am doing between 0 to 24 psi for moving air through the valve.

Okay...
take a look at valve data sheet
http://content.smcetech.com/pdf/PVQ.pdf
page 686 chart 1 is the transfer curve for your valve which by part# has 1.6mm orifice
would paste it if i could
it shows four different curves, one for each of four pressures
Curve D is for 0.2 mpa which i think is 29psi.
Observe valve doesn't begin to open until 140 ma at 29 psi
and at lower pressure, even higher current
So with discrete current steps of 20 ma , that'd sure resemble a "leap" at the last step.

Now i don't know if that's what is happening
but it deserves a look.

That the two valves behave differently suggests they have different internal friction

now go to page 684, section " working principle "
cutaway drawing of PVQ30
coil and core make an electromagnet that pulls UP on the armature to open the valve.
Armature is pulled down by gravity and pushed down by spring.
That's the balance that determines how far the valve is open, electromagnetic pull vs gravity and spring.
Almost.
There's one more UP force:
incoming air pressure underneath that black valve disc pushing up.
That's why it opens at different currents for different pressures.
Clearly the designers chose spring and area of orifice carefully to balance all the forces. That's what valve designers do.

If you take it apart don't be surprised if the disc is not flat on bottom but has a Cyrano deBergerac "nose" protruding down into the incoming air passage. I don't know if it will but that's a trick used on giant steam valves.

And that "up" force from air is why it could be unstable if current isn't controlled with authority. But we'll get to that later on.
For now, check very carefully those two valves... time how long it takes each to fill a garbage bag or balloon with air at known pressure and current. More air pressure will give more consistent results, as indicated by that family of curves on page 686.

Does this make sense ?

Objective now is to resolve difference in the two valves.
If it were mine and it indeed proved sticky, i'd consider cleaning it with rubbing alcohol or something. Don't use tap water it's too dirty. Microscopic grit in the sliding surfaces will be deadly.
Follow with a half hour air blow dry, half current to warm it.

See precaution about temperature, it should be kept below 100C.

What do you think ?

old jim

 

[Lexilighty]

What do you think ?

I think whatever you're taking, I want ALL of it! You are darn good!

I will try isopropyl alcohol on it tomorrow morning as I have none of that with me right now.

 

[jim hardy]

I will try isopropyl alcohol on it tomorrow morning

first try a little more air pressure...

thanks for the kind words, but your success is due your good observations and your perseverance .

 

[Lexilighty]

Yeah, after flushing for 30 minutes with air (this was after I had rinsed with isopropyl alcohol), I turned on the bad valve at 165mA and I hear a humming sound. Measured the DC Volts across OUT and +24V (which is constant btw, and I notice 17.41V DC and 4.351V a.c. Valve would not turn on. But when I apply 24V from power supply straight across its terminals, I can move air through it.
MOSFET is not hot. Just mildly warm.

 

[jim hardy]

Meaured the DC Volts across OUT and +24V (which is constant btw, and I notice 17.41V DC and 4.351V a.c. Valve would not turn on. But when I apply 24V from power supply straight across its terminals, I can move air through it.

hmmm solenoid will open if pulled hard enough.

Humming might be solenoid just dancing on bottom, relieving a little air with each bounce.

If your dmm has a frequency button see if you can get a number for that 4VAC.
120hz is power supply, something else low is likely solenoid's natural frequency.Does more air pressure straighten it out ?
How about cycling it a hunfdred times - is RIO programmable enough to handle FOR-NEXT loop ?
That diode in parallel with coil -see if making it into a diode plus 30 ohm resistor changes it.

Valve may be a candidate for post-mortem analysis.

 

[jim hardy]

17,41/165 = 105 ohms

does DMM agree ?
What's other valve read by DMM ?

Low resistance is indicative of shorted turns in coil.
Result of overheat probably.

 

[Lexilighty]

Does more air pressure straighten it out ?

I note that at higher air pressures, the valve will not open beyond 23-ish psi when i put 24V across its terminals.

17,41/165 = 105 ohms

Nope. DMM says its infinite impedance. Probably explains why armature is not lifted.

 

[Lexilighty]

With 24V across it and at any pressure more than 20psi, it will not open up. It opens with full 24V below 20 psi.

 

[Lexilighty]

Something interesting I have just noticed, by reversing the direction of air inlet, valve works beyond 20 psi with 24V across it. This is somewhat stated in the manual ( I just stumbled upon it this morning that pressure should be applied on the P1 port.

 

[Lexilighty]

Yes! Valve now works with proportionality. The trick was in the reversal of the inlet/outlet port.

Page 695 of Valve sheet:

Caution
When the product is used in vacuum, apply vacuum pressure to A(2) port.
The pressure at P(1) port should be larger than the pressure at A(2) port.

Science is but language well-arranged!

 

[jim hardy]

great news

Nope. DMM says its infinite impedance. Probably explains why armature is not lifted.

We ought to solve that mystery. Which valve was that, working or not working ?

The trick was in the reversal of the inlet/outlet port.

That's in interesting observation
Look at cutaway drawing on page 692
pressure on port P(1) pushes UP on armature with force P_P(1) X area of that middle part of disc covering P(1), call that area Area_P(1).
pressure on port A(1) pushes DOWN on the armature with force P_A(1) X (area of armature - Area_(P1))

net force on armature from air is difference of those two forces

you now have a number for milliamps to overcome that force at one pressure. It'd be fun to plot ma to just overcome backward dp vs pressure

Observe they chose direction of flow so increasing pressure across valve pushes it further open
which reduces pressure across it
helping it be stable
Backward pressure helps it be unstable - might that be the buzzing you heard?
\

Does the bad one read any ohms to its metal case? That'd confirm it's shot

Great news there Lex

you done good my friend !

 

[Lexilighty]

Observe they chose direction of flow so increasing pressure across valve pushes it further open
which reduces pressure across it
helping it be stable
Backward pressure helps it be unstable - might that be the buzzing you heard?
\

Again, great analysis!

Does the bad one read any ohms to its metal case? That'd confirm it's shot

I will check this on Sunday/Monday and revert to you. I have since left the lab.

 

[jim hardy]

I promised you a pointer to an interesting essay by Lavoisier

i stumbled across it in a library book and it affected my life. I kept a copy over my desk for decades.
here it is:
http://web.lemoyne.edu/giunta/EA/LAVPREFann.HTML

last paragraph:

the sciences have made progress, because philosophers have applied themselves with more attention to observe, and have communicated to their language that precision and accuracy which they have employed in their observations: In correcting their language they reason better."

Two paragraphs prior is an outrageously funny observation that you'll see demonstrated aplenty over the course of a career.
It begins "Instead of applying observation to the things we wished to know..."
Be careful how you use it around work for some don't like their foibles to be pointed out

I hope you enjoy him as much as i have.

old jim

 

[Lexilighty]

Thanks a lot for the reference. I will look through it.
Back to your former question, the valve still gives proportionality but I am still reading infinite impedance across it.

 

[jim hardy]

Back to your former question, the valve still gives proportionality but I am still reading infinite impedance across it.

well that's curious
it's certainly not open if it responds to current
some dmm's don't like to measure inductive loads
try a couple different scales?

of course you're measuring valve out of circuit
and you checked that your dmm reads less than an ohm when you short its own leads..oh well, you'll find it.

time to move on to making this thing do what it's intended to do - measure and control air flow?

I had fun, hope you did too ... and CONGRATULATIONS AGAIN !

keep us posted on your progress ?

old jim

 

[Lexilighty]

well that's curious
it's certainly not open if it responds to current
some dmm's don't like to measure inductive loads
try a couple different scales?

of course you're measuring valve out of circuit
and you checked that your dmm reads less than an ohm when you short its own leads..

Yes, you are right. It does work. I tried with a different DMM and this has been confirmed.

time to move on to making this thing do what it's intended to do - measure and control air flow?

Yes, now onto the control design for my pneumatic system.

Thanks a lot for your help. I hope our paths continue to cross and cross. :)

 

[jim hardy]

I hope our paths continue to cross and cross. :)

me too.

 

[Lexilighty]

Hey Old Jim,

Trust you are well. I am designing a variable voltage source, this time for a different proportional pneumatic valve that takes voltage between 0 to 30V and max current of 400mA. Specifically, this is for the Dakota Instruments valve with parts number 6ASV0105. The data sheet is https://www.dakotainstruments.com/d...al_6apsv01_proportionating_solenoid_valve.pdf.

Please do let me know if you have any pointers on how I can go from here.

 

[jim hardy]

lease do let me know if you have any pointers on how I can go from here.

Hi Lex

please forgive the delay

i have a big project underway, had the yard all dug up, and made a trip out of town as soon as we covered the trenches. Unanswered mail is waist deep. I had to pretty much set everything aside for a couple weeks.

Will take a look at your project this evening

old jim