[labnetwork] H2 Generators

Fri Jul 20 13:29:58 EDT 2012

Colleagues,

I'd like to add another note of caution. I too have worked 
with H2 for a considerable amount of time. Given most lab's 
volumes I do not agree a gas cabinet is "best practice" for 
180 cu ft cylinder of H2. Having said this I'd like to stay 
of out that fray (note the LEL's for H2/air are ~17%, the 
LFL's ~ 5%; it doesn't take a big room to dilute H2 to below 
these values.

Worth adding to this discussion is the routing of H2 lines 
and the danger of H2 mixtures in closed chambers.

H2, when routed through lines near a ceiling can pool H2 in 
dangerous amounts. For a mocvd installation, done some years 
ago in a room with a t-grid ceiling, an exhaust was added at 
the ceiling level to prevent the danger of pooling H2, a 
lighter gas than air.

A mixture of H2 and air or another oxidizer, in a closed 
chamber such as a furnace tube or bell-jar, is a significant 
hazard. The brisance of H2 reactions can result in 
significant force. Proper mixtures of Ar/H2 or N2/H2 
eliminate this risk. Ratios on the order of 20/1 are common. 
A few decades ago I routinely (daily) used forming gas (N2 
80%/H2 20%) in confined spaces (mircowave tube processing) 
with open flames. I never witnessed that ratio burn. That 
was best known method at that time.

Currently, the commonly accepted ratio H2 mixes is 5% in 
some other inert gas.

I agree RFO's are a great line of defense. As per John 
Shott/Stanford we use the DISS standard for all of our 
dangerous gases and incorporated DISS + RFO's early on.

Bob

PS as an added note, the worst lab explosion I've witnessed 
(after the fact) was an iced-over LN Dewar

Robert M. Hamilton
Marvel NanoLab
University of CA at Berkeley
Rm 520 Sutardja Dai Hall
Berkeley, CA 94720-1754
bob at eecs.berkeley.edu
(e-mail preferred)
510-809-8600
510-325-7557 (mobile - emergencies)

On 7/20/2012 7:34 AM, Weaver, John R wrote:
>
> John --
>
> My addled brain said it backward! Yes, the flow rate 
> allowed by the fixed orifice should be higher than the 
> trip point of the excess-flow sensor. The key is that you 
> need to allow the sensor to trip in a downstream failure 
> that causes excess flow while limiting the maximum 
> allowable flow.
>
> My apologies to the group for my confusing remark!
>
> John
>
> */John R. Weaver/*
>
> */Facility Manager/*
>
> */Birck Nanotechnology Center/*
>
> */Purdue University/*
>
> */(765) 494-5494/*
>
> */jrweaver at purdue.edu/*
>
> *From:*John Shott [mailto:shott at stanford.edu]
> *Sent:* Friday, July 20, 2012 10:23 AM
> *To:* labnetwork at mtl.mit.edu; John Weaver
> *Subject:* Re: [labnetwork] H2 Generators
>
> John et al:
>
> I am also a big believer in excess-flow switches and RFOs 
> in virtually all cylinders. We typically select a 0.010" 
> RFO unless there is a good reason to go larger or 
> smaller.  It is normally installed by the gas supplier in 
> the cylinder valve before it ever shows up at our site ... 
> although we do try to confirm that the proper RFO has been 
> installed before we install it.  However, I'm a bit 
> confused by the comment about setting the excess-flow 
> sensor ABOVE the flow rate allowed by the orifice .... 
> maybe I'm guilty of the mistake that you are describing.  
> If the excess flow switch is downstream of the RFO (which 
> it is in all cases for us), doesn't that mean that the 
> excess flow switch would never trip if it is set higher 
> than the RFO-limited flow (at least under steady-flow 
> conditions)?  We typically try to size our excess flow 
> switch so that it will trip at a flow that is 2-3 times 
> higher than the maximum expected flow but that is 
> typically several times LOWER than the RFO-limited flow 
> (at least at full cylinder pressure).  For most of our 
> gases, we would likely have excess flow switches set to 
> trip at 6-10 SLM.  Am I off the mark?
>
> More directly related to the hydrogen discussion, we've 
> recently changed our hydrogen sensors from LEL to 0-1000 
> ppm. While I believe that the legal requirement is to 
> detect a leak that reaches 25% LEL, all of our hydrogen 
> sensors are now set to give us a warning at 200 ppm and an 
> alarm at 400 ppm.  Our thinking is that we are better off 
> catching a leak as early as possible, since small leaks 
> seem to have a way of becoming large leaks.
>
> My apologies to all for not staying on the topic of H2 
> generators ... but whether coming from cylinders, 
> generators, or a bulk cryogenic source, safe handling and 
> distribution of H2 is important to us all.
>
> Thanks,
>
> John
>
> On 7/20/2012 5:51 AM, Weaver, John R wrote:
>
> I'm also a believer in excess-flow sensors and fixed 
> orifices. It seems obvious (but I have seen this mistake), 
> but the excess-flow sensor must be set above the flow rate 
> allowed by the orifice J.
>
>
>
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://www-mtl.mit.edu/mailman/listinfo.cgi/labnetwork

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20120720/76aad36e/attachment.html>