Setting up Mass Flow Controller

For posts specifically relating to fusor design, construction, and operation.
Post Reply
Tom McCarthy
Posts: 405
Joined: Wed May 08, 2013 3:36 pm
Real name: Tom McCarthy
Location: Ireland
Contact:

Setting up Mass Flow Controller

Post by Tom McCarthy »

I've been working on getting the mass flow controller usable. I bought it stand-alone, with no proper controller. It's a Hastings-Teledyne HFM-300: http://www.teledyne-hi.com/products-ser ... 00-hfc-302

For starters, I needed a +- 15V DC supply. I bought a block supply, XP Power ECL15UD02-s for ease of use - Just cut the head off a kettle cable and plug in the live and neutral mains leads to the screw terminals, and + and - 15v comes out the other side relative to a common.

I also needed a way to set the flow rate I want, which is done by inputting 0-5v to pin 14. I was going to just connect it to the PWM output of an Arduino and assumed that would work. However, some advice said I'd need to put the PWM from the Arduino through an RC filter and take a real analog 0-5 v from there. Using an oscilloscope to see which provided the best range and ripple, a 10k Ohm resistor and 10 microfarad resistor worked best. The control is actually split into two parts. One Arduino pin reads the 0-5v input from a potentiometer, and this produces a 0-5v volt output from another pin linked to said RC circuit. I used the potentiometer because it adds the human, intuitive side of fiddly-knob control that Richard is such a proponent of. Makes things much easier to get a feel for than inputting values on the laptop. I'm not sure how this will turn out in the final version though, as everything will eventually be remote controlled.

The final part is reading the 0-5v output from the MFC. This is just the given pin from the MFC connected to an analog input on the Arduino. The pin diagram from the manual (picture below) shows the MFC output being read between pins 5 and 6, but the Arduino pin is just connected to pin 6 and reads relative to ground. I reason that as all the commons in the power supply, Arduino, and MFC are connected together, the analog input reads relative to the Arduino's ground and the Arduino ground and pin 5 (signal common) are connected together and so the analog input is reading the MFC appropriately.

I tried the setup and it might need some tuning. Clamping the air compressor gun to the in side of the MFC, it showed a quick response in the out signal to any air coming through. A good blow from the compressor moved the MFC output straight to >5 volts (small bit uncalibrated my guess). Mind you, it didn't take much to put the flow rate at 100% - the MFC is rated for 10SCCM N2 so it's pretty sensitive. A problem that's appearing is the MFC doesn't seem to be actually controlling the mass flow, though it is reading it. I've the potentiometer and signal in set at 2v, but the MFC does not seem to be reducing the flow rate when I put the compressor through it, no reduction in signal out from >5v after a minute. Maybe I should wait longer. The MFC is lukewarm at this stage, I give it about an hour to heat up.

I made a slight mistake initially with the fitting for the D-sub 15 connection on the MFC. I initially ordered this http://ie.farnell.com/multicomp/8lcm015 ... eated=true

and quickly figured it was useless for soldering things on to. A vertical fitting with little solder receptacles for each pin was much easier to make use of: http://ie.farnell.com/multicomp/5501-09 ... dp/1084678

I'll try and get a more modest air supply than Dad's 1.5hP compressor to give me finer control over air in than all or nothing. Makes a ton of noise too, could only use it in the house because Mom was away...
mfc_close.jpg
power supply block upper left, breadboard, arduino and MFC
power supply block upper left, breadboard, arduino and MFC
potentiometer on bottom right of board, RC filter on middle bottom.
potentiometer on bottom right of board, RC filter on middle bottom.
User avatar
Garrett Young
Posts: 123
Joined: Fri May 26, 2006 9:51 pm
Real name: Garrett Young

Re: Setting up Mass Flow Controller

Post by Garrett Young »

HFM-300 is a meter and HFC-302 is a controller
- Garrett
Tom McCarthy
Posts: 405
Joined: Wed May 08, 2013 3:36 pm
Real name: Tom McCarthy
Location: Ireland
Contact:

Re: Setting up Mass Flow Controller

Post by Tom McCarthy »

Crap
Tom McCarthy
Posts: 405
Joined: Wed May 08, 2013 3:36 pm
Real name: Tom McCarthy
Location: Ireland
Contact:

Re: Setting up Mass Flow Controller

Post by Tom McCarthy »

Got a replacement mass flow controller. Vacuum General FV6-11. All Vacuum General information seems to have disappeared, luckily the seller found a pinout for it and sent it along, and Jerry BIehler found a similar model's pinout which is quite close to the one sent.

On the pinout there are a few things listed that I can't quite figure out. I only want to use the necessary connections - i.e power in, signal out and in, grounds. On the previous unit, not all connections were needed - for example, it functioned fine without the valve override pin in use as it would only come into effect if connected at ±15V. I am not sure all the additional pins on this MFC can be left unconnected, and I'm not sure what they do, so I need some help. Picture attached of the pinout.

DC VALVE RTN. (UV)
POS SHUT OFF (HI)
POS SHUT OFF (LO)
EXT ZERO

I suspect: It seems like the the pos shut off pins are overrides, as there is no override pin explicitly listed on the pinout, and there is (in some form) on most other MFCs I have seen. Perhaps, if either is powered then the MFC will be forced into the fully open (HI) or closed (LO) position. Could someone verify this? If so, it should function fine without them connected as they have to be positive for effect.

I am clueless about dc valve rtn. (uv). Google hasn't yielded anything. Does DC stand for directional control or direct current? I'm inclined towards directional control as all other mentions of direct current are VDC as opposed to DC. Also rtn = return? Then the directional control valve would prevent backflow, maybe when powered it will close, in that case it wouldn't need a connection.

Ext zero = external zero, I assume. Not sure what that's used for.

I'm going to try and run the MFC without these pins connected, and if it works all is good. I'd like to know how they're all used though.
MFC Pinout
MFC Pinout
Tom
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Setting up Mass Flow Controller

Post by Richard Hull »

What is the SCM value of the MFC?
I hope it is very low or control could be difficult.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Tom McCarthy
Posts: 405
Joined: Wed May 08, 2013 3:36 pm
Real name: Tom McCarthy
Location: Ireland
Contact:

Re: Setting up Mass Flow Controller

Post by Tom McCarthy »

10 SCCM N2, good for control. Also have a laser drilled orifice for a VCR fitting and relevant parts for flow choking too, but it's extra hassle and at the moment don't need it.

I heard back from the eBay seller: DC valve rtn is a ground for the override apparently.

Tom
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Setting up Mass Flow Controller

Post by John Futter »

Tom
careful with the compressor most controllers want a few psi to a max of 30psi

have fun
John Myers
Posts: 68
Joined: Tue Mar 07, 2017 7:13 pm
Real name:
Location: SoCal

Re: Setting up Mass Flow Controller

Post by John Myers »

I would guess the EXT ZERO pin is to set/calibrate the zero flow point.
Jerry Biehler
Posts: 975
Joined: Tue Nov 24, 2009 3:08 am
Real name:
Location: Beaverton, OR

Re: Setting up Mass Flow Controller

Post by Jerry Biehler »

There is no directional control in a MFC. It probably does not need to be hooked up just like the other three signals you have there.

Generally all you need is the +/-15v and the 5v for flow control. Some times you need the valve open signal which should just be a 5v signal and thats probably what the POS VALVE signals are for, hi is probably a signal to open the valve and lo closes. DC Return may need to be grounded, it may be the ref ground for valve control or it could be a signal out that tells the controller the valve state.

But first thing is to try is just add power and 5v signal and see if that gets it going.

When buying used stuff its a real good idea to see if you can get the manual before hand. A lot of these companies have been absorbed by others so it can be a real pain to find out who has the docs now. I cant find any MFCs with compatible pinouts.
Tom McCarthy
Posts: 405
Joined: Wed May 08, 2013 3:36 pm
Real name: Tom McCarthy
Location: Ireland
Contact:

Re: Setting up Mass Flow Controller

Post by Tom McCarthy »

Thanks for the help guys.

Hooked the MFC up with pretty much an identical setup to the original post here. Using pins ±15v, signal and power ground, and 0-5VDC In and 0-5VDC out. Using an Arduino, so reading range is 0-1023. Pos Shut offs (HI and LO) are grounded as are ext zero and dc valve rtn (UV).

MFC acknowledges flow, but is acting funky when controlling. I know it's a controller this time, and not just a meter because it says "MASS FLOW CONTROLLER" on the MFC. Normally, the reading from the MFC is 0, which is good. When I input air, reading immediately rises to say, 500 and then falls down to <100 and then single digits and then 0 within 3-4 seconds, while air is still going. Removing air in and doing same again quickly another one or two times, the high reading is less, and the fall to 0 much quicker, until the MFC doesn't seem to acknowledge flow at all. Seems like it might be trying to cut off flow altogether. Perhaps I need to connect one of the pins to a 5v high. Connecting Pos Shut Off (HI) does nothing.

Funny business.

Will keep at it, seller may have some more information, he seems pretty informed on these.

Tom
John Myers
Posts: 68
Joined: Tue Mar 07, 2017 7:13 pm
Real name:
Location: SoCal

Re: Setting up Mass Flow Controller

Post by John Myers »

I would leave the unused pins floating till you know if their active high or low signals.
Grounding the EXT ZERO may be forcing it to continuously try and zero out the flow reading.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Setting up Mass Flow Controller

Post by Richard Hull »

If I were testing an MFC I would leave the inlet open to air and hook the outlet to a vacuum chamber with a good gauge on it and go with a control signal from a simple 5 volt wipered potentiometer and watch the vacuum gauge as I adjust the pot. The Arduino, at this point, is a needless appendage and can be dealt with later provided the MFC works simply in a crude manner just described. Use the least complex and simplest scenario in tests. Once you have a proven device, you can add all sorts of fancy complications and then debug those with confidence.

I work with Arduinos a lot and love them, but I only apply them when I have a working device that needs control or feedback in a complex or rapid response, (sub-millisecond), time frame. I tend to design them into nuclear instrumentation or when I design multi-level control situations. There are a ton of Arduino like controllers out there, but few are as in-expensive to use and implement. Tiny PICs can be used when an Arduino is over kill and a propellar where you want to control large power plants and need the COGs. However, at $3.95 to a rip off price of $5.00 each, the Arduino pro-mini can wipe out the PIC at price and leave room for the "feature creature" to crawl into your design in future with unused ports and capabilities that you have left dormant.

I bought a MEGA 2650 for $24.00. I now consider this an insane moment in my life. I have yet to stress a pro mini to it's limits. The tons of ports and A to D on the mega just have not found a need in my world yet. The mega resides in my folly bin.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Jerry Biehler
Posts: 975
Joined: Tue Nov 24, 2009 3:08 am
Real name:
Location: Beaverton, OR

Re: Setting up Mass Flow Controller

Post by Jerry Biehler »

Yup, ditch the arduino and use a pot set up as a voltage divider. And do NOT just start randomly shorting pins. This is how you kill stuff. Leave everything but power and control in floating.
Tom McCarthy
Posts: 405
Joined: Wed May 08, 2013 3:36 pm
Real name: Tom McCarthy
Location: Ireland
Contact:

Re: Setting up Mass Flow Controller

Post by Tom McCarthy »

Thanks John, Jerry, Richard.

Lesson learned not to short pins without due consideration.

Set up as per Richard's instructions. Used a 10K pot and a variable 12v 2.5A supply for the voltage input. Set the supply at 5v across the pot. Watched output from MFC with a multimeter, a constant -160mV, with slight fluctuation. From the Pirani gauge on the chamber, pressure was 30 mbar ~ 20 microns.

With 5V input to the MFC, there was no current draw. As mentioned, output signal was a constant -160mV. Notice the negative output, should be in the range +0-5 V. ±15V was going in, and the PWR GROUND and SIGNAL GROUND pins were grounded to the commons of the 12V supply and ±15V supply.

The EXT ZERO, POS SHUT OFF (HI), POS SHUT OFF (LO), and DC VALVE RTN. (UV), CASE GROUND and duplicate PWR GROUND and SIGNAL GROUND pins were left floating.

Some calculations on detecting if the MFC is open with the vacuum gauge:
Chamber volume is 14 litres = 14000 cm3, 14000 SCCM at STP. At given pressure (30 mbar) 14 litres of air is equivalent to 0.00003 * 14000 SCCM ... About 0.5 SCCM. So, if I have the MFC fully open, giving a flow of 10 SCCM, I'd expect the pressure to rise about 20 times in the space of a minute to 600 mbar. However there is no such large rise, and instead a very slow leak, of 10 microns/minute.

How to proceed from here? MFC seems to be absolutely closed to any flow and the -160mV output is odd. Perhaps it requires another pin to be grounded/powered to open up. I'd rather not kill the thing, though it may already be gone. Can open it up for a look to see if any components are obviously crapped out.

I've also contacted Brooks Instruments, who provide support for legacy products from Tylan General, Unit, Millipore etc. In 2009 Brooks acquired Celerity which included Millipore spin off Mykrolis which in turn included Tylan General and so Vacuum General. Guy who deals with the old stuff is in office next week, may be able to help with information.

Tom
Jerry Biehler
Posts: 975
Joined: Tue Nov 24, 2009 3:08 am
Real name:
Location: Beaverton, OR

Re: Setting up Mass Flow Controller

Post by Jerry Biehler »

Ground pins must be grounded.
Tom McCarthy
Posts: 405
Joined: Wed May 08, 2013 3:36 pm
Real name: Tom McCarthy
Location: Ireland
Contact:

Re: Setting up Mass Flow Controller

Post by Tom McCarthy »

Grounded all the ground pins, same result, had no effect as far as I can tell.

I think it might be best to cut my losses, both time and money, and get a variable leak valve, use the laser orifice I've already got and I can even measure flowrate with the flowmeter previously described. Brooks might get back to me with a manual to confirm pinout, but the likelihood of that seems low.

Tom
Post Reply

Return to “Fusor Construction & Operation (& FAQs)”