Vapor Detection - Alarm / Exhaust Fan Control

[HedgeBird]

Looking for suggestions on alcohol vapor detectors/alarms.  Ideally trying to find something that can automatically switch on an exhaust fan once a certain concentration of alcohol vapor is detected. Also curious if anyone has used something more sophisticated that can tie into an existing alarm panel (that monitors the smoke/heat/sprinkler vales) and would send an alarm or supervisory notice to my monitoring company at certain concentrations of alcohol.

Also interested in simple detectors that just sound an audible alarm

Thanks!

[HedgeBird]

Just found this previous thread that seems to point me in the right directions..

Looks like the RKI Ps2 might be what I need.

[Jwalsh77]

If you remember, post what you ended up going with! I am in the process of setting up a distillery and this is one of my items to explore.  

[vsaks]

RKI PS2 with exhaust fans is the way to go.
Just be sure to calibrate them every six months or so. It's a relatively simple process outlined in the manual.

On ‎5‎/‎14‎/‎2019 at 8:57 AM, HedgeBird said:

Just found this previous thread that seems to point me in the right directions..

Looks like the RKI Ps2 might be what I need.

[bluestar]

While more expensive alarms like the RKI PS2 are fully certified and calibrated, it is not clear they will be any more useful than a cheaper combustible gas LEL detector. There are many on the market today meant to detect methane/butane/propane. The cross sensitivity of the detectors used for these between ethanol and butane or methane is close to 1, hence they will likely detect ethanol vapor at the same levels as these other gases. Also, the LEL for these gases ranges from 1% to 5%, and ethanol is 3.3%, so an alarm designed to go off at 10% LEL for general combustible gas is likely to be suitable for use with ethanol. Some of these can be purchases for less than $20, and in addition to an audible and visible alarm, may have a relay output. Is there any reason why one or more of these can not be used instead of something like the RKI PS2 that costs $400+?

[Thatch]

31 minutes ago, bluestar said:

Is there any reason why one or more of these can not be used instead of something like the RKI PS2 that costs $400+?

Yes, what does your insurance company and AHJ mandate.  Ours liked the RKI and fans turning on.  We also have audible and visual alarms.  Your insurance company and AHJ may not care.  Sometimes this is not about logic but instead about dotting the i's and crossing the t's.

[starcat]

The Architect Spec'd this model Q8, so this is what we ended up with:

http://www.qelsafety.com/wp-content/uploads/2012/09/86350-001-000-RA-Q8-Operation-Manual-1.pdf

There are 3 SPDT relays on board that can be configured to stop process and so forth.

[vsaks]

You can use another cheaper alarm if
1. The alarm can be calibrated
2. The range is suitable for your use
3. The sensor being used is stable enough that it doesn't drift
Even the RKI PS2 needs to be periodically calibrated. We used 2000 ppm Hydrogen and 4000 ppm Hydrogen as surrogates to calibrate at 10% and 20% LEL of ethanol as recommended by RKI (though my research indicated that it should have been 2800 ppm and 5600 ppm Hydrogen).

You need to match the LEL equivalence. 10% LEL of ethanol is the same as 9% LEL methane, 13% LEL propane, or 14% LEL n-butane. Double those for 20% LEL ethanol.
An off the shelf alarm at 5% of methane will trigger at 6% LEL ethanol. Not sure if you really want that.
 

1 hour ago, bluestar said:

While more expensive alarms like the RKI PS2 are fully certified and calibrated, it is not clear they will be any more useful than a cheaper combustible gas LEL detector. There are many on the market today meant to detect methane/butane/propane. The cross sensitivity of the detectors used for these between ethanol and butane or methane is close to 1, hence they will likely detect ethanol vapor at the same levels as these other gases. Also, the LEL for these gases ranges from 1% to 5%, and ethanol is 3.3%, so an alarm designed to go off at 10% LEL for general combustible gas is likely to be suitable for use with ethanol. Some of these can be purchases for less than $20, and in addition to an audible and visible alarm, may have a relay output. Is there any reason why one or more of these can not be used instead of something like the RKI PS2 that costs $400+?

[bluestar]

21 hours ago, Thatch said:

Yes, what does your insurance company and AHJ mandate.  Ours liked the RKI and fans turning on.  We also have audible and visual alarms.  Your insurance company and AHJ may not care.  Sometimes this is not about logic but instead about dotting the i's and crossing the t's.

As I said, the difference is certification. Many of these cheaper units have audible alarm, visual alarm, relay for turning fans on, and some even display the nominal ppm or %LEL of the combustible gas. Note these units have come onto the market in the past couple years for use in homes and RVs for fuel gas detection. Most are Chinese manufacture. Like other consumer versions, rather than replace the sensor periodically (which you must do as part of the maintenance and calibration of the more expensive units), you replace the whole unit in most cases every couple years.

[bluestar]

20 hours ago, vsaks said:

You can use another cheaper alarm if
1. The alarm can be calibrated
2. The range is suitable for your use
3. The sensor being used is stable enough that it doesn't drift
Even the RKI PS2 needs to be periodically calibrated. We used 2000 ppm Hydrogen and 4000 ppm Hydrogen as surrogates to calibrate at 10% and 20% LEL of ethanol as recommended by RKI (though my research indicated that it should have been 2800 ppm and 5600 ppm Hydrogen).

You need to match the LEL equivalence. 10% LEL of ethanol is the same as 9% LEL methane, 13% LEL propane, or 14% LEL n-butane. Double those for 20% LEL ethanol.
An off the shelf alarm at 5% of methane will trigger at 6% LEL ethanol. Not sure if you really want that.
 

Everything you said is correct. But as you note, everything needs to be calibrated, or nothing. If you get a unit that can read out ppm or %LEL, you can calibrate. What would be wrong with triggering at 6% LEL ethanol, if 10% is required? You are not prevented from being more stringent. Note, I agree, if you can afford the fancier units, it is a better choice, and if your FM or insurance company requires it, then there you go. But wouldn't it be better for many small distilleries (vast majority) operating without any combustible sensors be operating with what is easily and cheaply available?

[bluestar]

Oh, I should add, some of the more expensive units are also explosion proof for this application. Which means, the cost to properly install them is also high, since they will need to be wired in using contained solid conduit meeting those requirements. And, if you are locating the sensor where everything else is required to be explosion proof (like your lights, electrical outlets, power and control wiring), then certainly you want your sensor to be explosion proof as well. If you want to put the sensor close to the still, for example, this will likely be the case. Probably not an issue if being used in a storage area for barrels. But if used near the still, you want to use it to do more than turn on a fan: you should also have it failsafe to kill heating power to the still. I don't think the RKI PS2 is explosion proof, so it would not qualify. RKI does make remote sensors that are explosion proof, akin to the Q8 mentioned elsewhere by @starcat.

Finally, original poster asked about tying into fire alarm. Again, there are a fair number of less expensive units available for that now (although more expensive than the cheap home units), offered by the same companies that sell the smoke and CO alarms currently used in fire alarm installations, and for the same purpose: to detect combustible gas. In some areas, this is becoming a requirement anywhere that gas-fired heaters, boilers, ovens, and stoves are located.

[vsaks]

5 hours ago, bluestar said:

Everything you said is correct. But as you note, everything needs to be calibrated, or nothing. If you get a unit that can read out ppm or %LEL, you can calibrate. What would be wrong with triggering at 6% LEL ethanol, if 10% is required? You are not prevented from being more stringent. Note, I agree, if you can afford the fancier units, it is a better choice, and if your FM or insurance company requires it, then there you go. But wouldn't it be better for many small distilleries (vast majority) operating without any combustible sensors be operating with what is easily and cheaply available?

There is no danger until you hit LEL. So a lower limit gives you early warning to fix whatever is wrong, or to get the hell out of there. We chose 10% and 20%, not because it was required by AHJ, but because we felt it gave us enough headroom. If you have an alarm which gives you LEL readout and you can choose the actionable threshold, that should allow you to run operations safely (now whether your AHJ accepts that or not is another issue).
The reason you don't want a 6% or lower threshold is you don't want a threshold so low that you hit it during normal operations. Then it just becomes a nuisance and might not be noticed when a real problem is happening. Our $450 RKI sensors needed calibration after 6 months or so. They were being triggered multiple times a day. No alarms since the calibration. That illustrates two things :
1. All gas sensor based alarms need periodic recalibration. If the AHJ sees the alarm installed, they are going to ask for calibration records.
2. I don't know that the alarm threshold should be, but do know that when our alarms dipped below 10%. it was too low
 

[Silk City Distillers]

11 hours ago, vsaks said:

The reason you don't want a 6% or lower threshold is you don't want a threshold so low that you hit it during normal operations. Then it just becomes a nuisance and might not be noticed when a real problem is happening.

Surprised you would even hit a "real" 6% LEL, that's actually a very high reading.  Of course, it depends on the size of your space, the air changes per hour, etc.  Yes, I realize that's 6% of the 3.3% LEL of ethanol vapor.  Suspect that meter drift gave you an inaccurately high baseline reading (zero calibration).

Ethanol vapor would be detectable by odor at about 85-100ppm, and 1000ppm is the OSHA exposure limit.

6% LEL would be approximately 2000ppm, twice the legal exposure limit.

3300ppm would be the 10% LEL, 6600ppm would be 20% LEL.

By 5000ppm (15% LEL), ethanol vapor would begin to be uncomfortable, by 10,000 (30% LEL) there would be obvious discomfort, by 15,000 (50% LEL) you would be continuously coughing and tearing. 20,000ppm (70% LEL) is completely intolerable.

Previously, 15,000ppm was the NIOSH IDLH limit - Immediately Dangerous to Life or Health.  But that's been revised down to 3,300ppm (10% LEL).  This is likely the reason an engineer or AHJ would specify 10% LEL alarms - that limit is about people safety, not explosion risk safety.

The other factor to keep in mind, is that localized readings might be higher.  A still leaking vapor could have a much higher concentration at the still, and might not be so obviously noticeable otherwise.  Just being near 15% LEL, would likely be very noticeable to a skilled distiller, you are your own combustible gas detector.  But you might not be close enough to notice, until it was too late.

[Silk City Distillers]

We are using a Safe-T-Net 410 monitor control panel with two Drager Polytron XP combustible gas sensors.  One behind the still, one next to gauging/blending.  They are connected via XP rigid conduit, seal glands, the whole 9 yards.  The control panel has two more sensor inputs, I two more sensors but I haven't installed them.

You can find similar panels on eBay for relatively low prices.  Most of these panels will expect a 4-20ma sensor input, and will have a variety of alarm and relay output options.  Ours is currently connected to a siren, but it would be very easy for us to connect the panel to our still controller, and shut off the boiler, power, turn on fans, etc.

You can find high quality explosion proof detectors inexpensively as well, but your mileage may vary.  You absolutely need to understand how to program and calibrate these things, otherwise they are just paperweights.  If you buy a used one, you might find that it's no longer able to be calibrated, so it's a gamble.

This is absolutely NOT DIY beginner territory.  You'll probably waste tons of money if you even attempt this, and end up wasting even more time trying to educate yourself on it.

Sensors to look for would be Drager, MSA, RKI, Bacharach, etc.  The minute you go into explosion proof territory, you pay a serious premium.

[Silk City Distillers]

I'll post this here.  I have no liability, offer no warranty or guarantee, no connection or affiliation with the seller.

https://www.ebay.com/itm/MSA-Ultima-XIR-Gas-Monitor-NEW/264294563430?hash=item3d892efe66:g:LncAAOSwA8Zcvk0l

If it works, which it might not, it's pretty much the best of the best, infrared combustible gas detection, very very nice unit.  Probably sell for $2,500-3,000 new.  Combustible gas sensor, calibrated with methane.  Separate remote sensor and controller head.  These are XP and your electrician can pull XP rigid conduit.  Plenty of relay outputs.  At 240 bucks, even if your electrician charges you $1000 to pull explosion proof conduit, you would be ahead of the game.  This is the kind of sensor you could feel very comfortable with locating at the still.

 

[HedgeBird]

Hey Silk - Tried sending you a msg, but not able to via the forum email system.  Would love to touch base on the phone if you have time...

[Silk City Distillers]

@HedgeBird - Cleaned it up.

Some photos.

[vsaks]

Did not know about OSHA stuff. That makes a lot of sense. Thanks for this information.

On ‎5‎/‎17‎/‎2019 at 3:31 AM, Silk City Distillers said:

Surprised you would even hit a "real" 6% LEL, that's actually a very high reading.  Of course, it depends on the size of your space, the air changes per hour, etc.  Yes, I realize that's 6% of the 3.3% LEL of ethanol vapor.  Suspect that meter drift gave you an inaccurately high baseline reading (zero calibration).

Ethanol vapor would be detectable by odor at about 85-100ppm, and 1000ppm is the OSHA exposure limit.

6% LEL would be approximately 2000ppm, twice the legal exposure limit.

3300ppm would be the 10% LEL, 6600ppm would be 20% LEL.

By 5000ppm (15% LEL), ethanol vapor would begin to be uncomfortable, by 10,000 (30% LEL) there would be obvious discomfort, by 15,000 (50% LEL) you would be continuously coughing and tearing. 20,000ppm (70% LEL) is completely intolerable.

Previously, 15,000ppm was the NIOSH IDLH limit - Immediately Dangerous to Life or Health.  But that's been revised down to 3,300ppm (10% LEL).  This is likely the reason an engineer or AHJ would specify 10% LEL alarms - that limit is about people safety, not explosion risk safety.

The other factor to keep in mind, is that localized readings might be higher.  A still leaking vapor could have a much higher concentration at the still, and might not be so obviously noticeable otherwise.  Just being near 15% LEL, would likely be very noticeable to a skilled distiller, you are your own combustible gas detector.  But you might not be close enough to notice, until it was too late.

[Pour Decisions]

Older thread, I know.... Those with the RKI PS2 unit, is there a recommended location for it to be mounted? Height from floor would need to make a difference as well?

[Thatch]

9 hours ago, Madspeed said:

Height from floor would need to make a difference as well?

We have ours mounted on a pillar at eye level but the sensor is somewhere between 6 to 12 inches from the floor.  We are within a few feet of our stills with the sensor.

[Pour Decisions]

Thanks Thatch. I was under the impression the sensor should be less than 18" from floor level, glad to hear that must be correct. Makes sense to place the main unit somewhere convenient and easily accessible.

 

[Thatch]

4 hours ago, Madspeed said:

should be less than 18" from floor level, glad to hear that must be correct.

It has to be low, keep in mind that ethanol vapor is heavier than air.  The only way to detect that you have a problem is to get the sensor low but not so low that you'll drown it in a washdown.