Mass flow controller question

For posts specifically relating to fusor design, construction, and operation.
Post Reply
brett_kvo
Posts: 26
Joined: Thu Jan 14, 2021 3:02 pm
Real name: Brett Kvo

Mass flow controller question

Post by brett_kvo »

Hey all,

We're trying to find an adjustable MFC that will work for our fusor. I am having a hard time finding info regarding many eBay MFCs, and I'm appealing to the collective wisdom of this group for some advice.

Do you think this MFC is variable from 0-20 SCCM? Or is it just 20 SCCM flat? https://www.ebay.com/itm/Unit-Instrumen ... SwhPZdf81g

Do you have any recommendations on specific MFCs that have worked well for you? We're just trying to minimize screw-ups, and buying the right MFC (without proper documentation on many of the eBay options) has proven challenging.

Thank you!
-Brett @ www.upsreactor.com
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Mass flow controller question

Post by Joe Gayo »

That's a meter not controller. Besides the label, you can tell because there's no proportional control valve.
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Mass flow controller question

Post by Joe Gayo »

Many will say, use small ID tubing and a needle valve.

I use a proportional valve, pressure gauge, and closed loop controller for pressure regulation (you can buy or make the electronics pretty simply). It's pretty great, to set the pressure and have it stay constant.
User avatar
Artem Artemov
Posts: 43
Joined: Sat Sep 21, 2019 4:57 pm
Real name: Artem Artemov
Location: Moscow, Soviet Union

Re: Mass flow controller question

Post by Artem Artemov »

It's not mfc, it's mfm. It is not suitable for gas control.

1. Choose an MFC with analog control (0-5 volt voltage change). This is the simplest thing. You don't need devicenet, I think.
2. The type of connection. Do you have vcr 1/4 fittings? The simplest is swagelok, you can even connect a teflon tube od 6.3 mm. VCR fittings are expensive.
3. I think that sccm d2 depends on the volume of the vacuum chamber. Usually, MFC works in the range of 2%-100%. If the chamber is medium or large, 20sccm is a good option.
Attachments
282CA169-8983-4337-8AB6-B2E950A13872.jpeg
User avatar
Artem Artemov
Posts: 43
Joined: Sat Sep 21, 2019 4:57 pm
Real name: Artem Artemov
Location: Moscow, Soviet Union

Re: Mass flow controller question

Post by Artem Artemov »

Sorry. I saw on your website that you have vcr fittings. You have excellent veriflo 944 valves! At the inlet there should be a veriflo valve , then a pressure regulator, then an mfc, then (optional) you can put a pneumatic valve, then a tube, and then another veriflo valve on the vacuum chamber. I would not mount the regulator and mfc on the vacuum chamber, but install the gas control in another place, connecting it to the chamber with a hose or capillary tube.

These regulator may not be designed for vacuum applications. And veriflo 944 series can be used for vacuum and they have a metal membrane. All equipment should be easily removed without breaking the tightness of the chamber, so the valve should stand first.
Attachments
70AB1732-22E0-40EC-85AB-BD61F6D2A2EC.jpeg
User avatar
Rich Feldman
Posts: 1470
Joined: Mon Dec 21, 2009 6:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Mass flow controller question

Post by Rich Feldman »

I've been happy with several purchases from Svcstore, generally picking them up in person since they are not far away. I think that's where I got my NIM bin and detector-bias power supply.

Here's a post by Joe Gayo about MFC's, with a followup by me about what's inside one I was given by Greg Courville.

viewtopic.php?f=18&t=12560&p=81611&#p81611
DSCN1018.JPG
Been meaning to follow up with characterization of the proportional valve.

It was pointed out that flowmetering based on heat transport is not very sensitive to gas type.
But proportional valve operation varies with the square root of molecular weight.
A valve sized for flow control up to 20 ccm of N2 would deliver more than 50 ccm of D2 with same solenoid current.
But I guess a MFC labeled for 5 or 10 sccm N2 might peg its meter before delivering 20 sccm D2. In my inexperienced opinion.
All models are wrong; some models are useful. -- George Box
brett_kvo
Posts: 26
Joined: Thu Jan 14, 2021 3:02 pm
Real name: Brett Kvo

Re: Mass flow controller question

Post by brett_kvo »

Wow! Thank you all for the thoughtful replies. For some reason, I didn't receive any reply notifications so I'm just seeing these now. We have our class this evening so I'll show this post to the students and we'll go from there!
Lukas Karoly
Posts: 1
Joined: Tue Feb 16, 2021 8:36 pm
Real name: Lukas Karoly

Re: Mass flow controller question

Post by Lukas Karoly »

Howdy, I'm Lukas, one of the students working with Brett.

I found this listing for an MFC on eBay (not an MFM, made to sure to read the label this time!):
https://www.ebay.com/itm/Unit-UFC-1000- ... Swtypc5uxP

Is this more what we what? It looks like the connections are the same vcr 1/4 fittings which would work with our set-up, is that correct? I am also not 100% confident on the whole calibration deal. It says 15 sccm N2, and from what I can find online, flow rates for thermal-based MFCs are the same for N2 and D2. Are there non-thermal-based MFCs, or can we be pretty confident that the flow range will be true for D2?

Cheers!
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: Mass flow controller question

Post by Liam David »

You're correct, those are 1/4" VCR fittings. Most MFCs use a thermal sensor, so the measured flow rate is dependent on the structure of the gas molecule. Nitrogen and deuterium are both diatomic, hence they have (almost) the same molar heat capacity. Per Rich Feldman's comment that the flow rate ~1/sqrt(m), the MFC has to close the valve more for D2 than it would for an equivalent flow of N2. MFC's don't necessarily regulate well near the bottom few percent of their range, which might be exacerbated by this. From my experience with a very similar MFC (Unit UFC-1020A, 20sccm N2), I couldn't get reliable, repeatable control below ~5%, and the response time from 0-x% flow was slow (>5s) and increased with a lower setpoint. There was a lot of hysteresis and to get <1sccm flows I had to approach it from above. Your linked MFC might have a completely different response time--the datasheets just aren't easy to find and I'm mentioning things you may encounter. If you purchase it, you probably need to flow >1sccm for stability, which for your large chamber would be expected anyways. Long story short, it'll probably work ok, there are just several unknowns.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Mass flow controller question

Post by Richard Hull »

Liam tells it like it is. I have several MFCs and due to the issues he brings up, I have refused to use them. I have found the most fabulous gas control in a precision leak valve first used only 1 year ago on fusor V and will never use any gas flow device but this one ever again. It is the cat's pajamas. I bought it for $50.00 at HEAS 3 years ago.

For those not familiar with what a precision leak looks like, I attach images with captions.

Richard Hull
Attachments
This is what $50 at HEAS 2017 bought....The leak is seen attached to a lot of other great goodies as a &quot;ride along&quot; of vacuum goodies.
This is what $50 at HEAS 2017 bought....The leak is seen attached to a lot of other great goodies as a "ride along" of vacuum goodies.
This another sapphire Leak I bought at HEAS 2019.  A bit too delicate of operation for my ham fisted manner.  It may be used in future.
This another sapphire Leak I bought at HEAS 2019. A bit too delicate of operation for my ham fisted manner. It may be used in future.
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
brett_kvo
Posts: 26
Joined: Thu Jan 14, 2021 3:02 pm
Real name: Brett Kvo

Re: Mass flow controller question

Post by brett_kvo »

Thanks for the advice all.

RIchard--any idea how you might digitally control one of those leak valves? I appreciate the precision and repeatability of it, but we're attempting full computer controls and the MFC checks the box.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Mass flow controller question

Post by Richard Hull »

Easy if you have the skills. Remove the knob and attach a high torque stepper motor. Count steps from off to various calibrated running gas points and interpolate between step counts to create an array that can step to numeric gas pressure to needed step counts. Or simply avoid this and have a forward and reverse rocker switch to drive the motor in or out. Finally and most desirable is that the motor stepper could be linked to the pressure gauge in software to cause it to seek a punched in pressure and hold it there in a virtual servo loop. Many ways to skin this cat once a stepper is attached.

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
brett_kvo
Posts: 26
Joined: Thu Jan 14, 2021 3:02 pm
Real name: Brett Kvo

Re: Mass flow controller question

Post by brett_kvo »

Great. Thank you. That's what I had in mind as well. I like this direction over the MFC. :)
User avatar
Bob Reite
Posts: 576
Joined: Sun Aug 25, 2013 9:03 pm
Real name: Bob Reite
Location: Wilkes Barre/Scranton area

Re: Mass flow controller question

Post by Bob Reite »

I use a 0-10 SCCM mass flow controller. I set the flow rate with a 10 turn pot. For my chamber, I usually operate between 3-5 SCCM.
The more reactive the materials, the more spectacular the failures.
The testing isn't over until the prototype is destroyed.
User avatar
Artem Artemov
Posts: 43
Joined: Sat Sep 21, 2019 4:57 pm
Real name: Artem Artemov
Location: Moscow, Soviet Union

Re: Mass flow controller question

Post by Artem Artemov »

I had an idea to connect a stepper motor to my needle valve, but this is archaic. You need an motor, a driver, a controller, calibration, and dance with drum. I think it's easier to use mfc with an analog control signal. But if you are serious guys, you can use mfs and vacuum gauge with devicenet, combining everything into one network. Many phillips vacuum gauges have a devicenet. Then you will need to buy an mfc with this function for perspective. You can fully automate everything.

In the future, I want to connect everything through this network. I bought vacuum gauge with devicenet, but they can be used without using this network, the gauge have screens. But such gauge with a screen are expensive(


Ps You can also use the MFC and needle valve at the same time. You give the signal to fully open the mfc and work manually with the valve wheel.
Attachments
FB2058C5-43FB-484F-9CDE-39052745F16F.jpeg
590884C2-A5D6-47E2-9525-D9EF0574A67C.jpeg
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: Mass flow controller question

Post by Liam David »

The Devicenet connector pinouts include can_h and can_l. If I'm not mistaken, you might be able to communicate with the gauge using the standard CAN bus. There are Arduino shields/breakouts that would make this pretty easy. I have a convectron gauge identical to Artem's, but haven't investigated whether this is possible and am not familiar with the standards, so take it as is.
brett_kvo
Posts: 26
Joined: Thu Jan 14, 2021 3:02 pm
Real name: Brett Kvo

Re: Mass flow controller question

Post by brett_kvo »

Hey Bob, could you tell me the exact make/model of your MFC? Just curious what's worked well for you.

We managed to find an old precision leak valve laying around, so I think we may go that route for version 1.0 (manual control) of our reactor. I think we'll pursue either an MFC or a stepper-motor driven leak valve moving forward.

Thank you all for your thoughtful responses. This information just can't be readily found anywhere else on the internet.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Mass flow controller question

Post by Richard Hull »

You are right Brett. We have years of accumulated knowledge here and lots of folks who have done the same thing in many different ways, successfully. You came to the right place. Thanks for the accolades to all of us and this site.

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
User avatar
Bob Reite
Posts: 576
Joined: Sun Aug 25, 2013 9:03 pm
Real name: Bob Reite
Location: Wilkes Barre/Scranton area

Re: Mass flow controller question

Post by Bob Reite »

Mine is a Sierra Instrumentation "Side track III" "cust code I-458 Range 0-10 SCCM Orifice .010" Gas H2. I found it on Ebay. Any MFC that can be set for 0-10 SCCM will do.
The more reactive the materials, the more spectacular the failures.
The testing isn't over until the prototype is destroyed.
brett_kvo
Posts: 26
Joined: Thu Jan 14, 2021 3:02 pm
Real name: Brett Kvo

Re: Mass flow controller question

Post by brett_kvo »

Thanks Bob! Much appreciated. :)
User avatar
russssellcrow
Posts: 68
Joined: Fri Dec 27, 2019 12:40 am
Real name: Russell Crow
Contact:

Re: Mass flow controller question

Post by russssellcrow »

We use a STEC SEC-7440M mass flow controller. It is older (Analog), rated at 200 SCCM, and calibrated in Argon.

We are able to control the chamber pressure in increments of 1 micron, from 10-200 microns. These beat the heck out of trying to balance pressure using a needle valve, the control is so much finer.
150px-DMS_Fusor_MFC.jpg
150px-DMS_Fusor_MFC.jpg (7.32 KiB) Viewed 4022 times
Post Reply

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