A Flash in the Can..MkII

[Touchwood]

Hi all,


A while back I posted some pics of the little LED flasher I built to put in my dust collector cyclone separator trash can. This gadget works fine for me except when I'm hogging off lots of material with the planer (like with soft maple or cypress) and the big shavings get stuck between the lid and the cyclone baffle. I never see anything blocking the LED because it's all up above in a big mess..with the shavings being dumped into the Grizz bottom bag, thus defeating the whole purpose of everything .:no:


The solution to this is to widen the throat of the baffle drop zone..which I did..and now all is fine. But a visitor to our site in Connecticut saw the "flash in the can" bit,(he is now actually a member) andi asked my advice on building a more elaborate sensor.
Folks with large cyclones (like Oneida) have seen the dust bin fill completely to the top and totally load the cyclone right up to the blower..these collectors are very strong and keep right on”sucking”somehow, even when no more chips can get in there. You'd think the motor would sound different, but I'm told it doesn't.
The guy in Connecticut has one of these and was trying to put some kind of sensor together to prevent possible overload. Oneida sells a sensor kit for $269 to monitor dust level that you have to wire it up to the 220V supply and install yourself.


http://store.oneida-air.com/levelmetersensor-prewiredkit.aspx


I suggested using the little photocell detectors used in electric garage door safety trips. The idea is to mount them on opposite sides of the drum at windows set at the fill level wanted. We also needed some delay so that random chips passing by don't trip the alarm, so a 5 second delay was designed into the circuit..in other words, the sending sensor has to be blocked continuously for 5 seconds before tripping.:widea:
Garage doors stop instantly if the beam is broken for a second or less.

Anyhow here's what we developed. The garage door sensors are Genie model STB-BL replacement units that you can get at most HD or Lowe's stores for about $35.00 . The circuit is packaged in a weatherproof (dust-proof) outdoor enclosure and a very bright strobe light is used for the alarm.

Here's the sensors looking at each other on the bench

You know when they're aligned..the sender glows green, the receiver glows red

Here's the box put together with the strobe light on top

...and here's what happens when the beam is interrupted for 5 seconds..

​

The strobe will flash continuously about once a second until the obstruction is removed..or you switch the power off. The whole thing is powered by a 12 volt "wall-wart", and the relay driving the strobe (which only draws a quarter-amp from the circuit) can supply at least 7 amps. So driving the magnetic switch on the dust collector to turn it off when the alarm lights is no problem at all.:icon_thum


We had the circuit assembled by Jeff Wheat at SimpleCircuitBoards.com,and he will sell the complete circuit board, populated with the components and tested, for $20.00 plus $6.00 Shipping. The strobe is from E-Bay and cost $9.97 delivered.



So for a total of about 70 bucks you have basically something that does pretty much the same thing as the Oneida sensor.:icon_cheers



I haven't tested this in real life yet, and it's way overkill for my little separators so I need a "lab rat" somewhere close by so I can help install it and watch it blow up.:nah:


I'm thinking Travis might want to try it out since he's already experienced a cyclone plug. Scott isn't up and running yet but this would work for him too.


Any takers?? :gar-La;

PS: I went into the shop this afternoon to get something and threw my baseball cap on the bench. ..without realizing it had landed between the sensors. Five seconds later it let me know !!!


Don
​

​

 

[Bas]

Don, this looks great. But, to really make an impression, you need to add the HF 12V Ooga horn into the system....

Glad to hear widening the drop zone solved your problem. Have you noticed any degradation of the separation (especially the fines)?

 

[McRabbet]

Don,

Your system is intriguing and looks like it would be effective to let you know when the sensors are blocked and it is time to empty the collection bag or bin. But why not make it automatic and have it shut the DC down when the alarm condition is sensed? Alan in Little Washington designed a system for his DC several years ago and I did an adaptation that uses a stock Dusk-to-Dawn photocell (~$12 at Lowe's) that triggers an inexpensive Allen-Bradley "Ice Cube" relay and controls the power relay to the DC and triggers an alarm light as well. I documented it in a thread here more than a year ago and built a prototype for my ClearVue Cyclone. I updated the circuit with a minor revision to provide a service switch (My DC is below my shop on a pad in a tall dry crawl space and I need to go around to an outside door for access to it from my shop). Here's the revised circuit (I now provide power from the remote switch to the candelabra lamp and photocell to simplify the system, but haven't updated the circuit diagram).

In the first picture, 7W Candelabra at top will sit one side of Bin mouth and shines on dusk-to-dawn photoeye in lower handy box. Photoeye stays ''off'' as long as the beam shine through to opposite side. Switch on top handy box controls power to bin light and sensor. This picture, (flash in darkened room) shows unit in normal operating mode.

When the light beam to the photoeye is blocked by sawdust in a full bin (here it is done artificially), the photoeye switches power to its Red output and energizes the ''Ice Cube'' relay coil, tripping the normally open contacts to the closed position and opening the contacts carrying power to the 230 V Contactor. The DC turns off and an indicator light is turned on.

One important note, the photocell sensor selection is important. Many (including most at Lowe's on the electrical component aisle) have a built-in delay of 15-25 seconds to avoid premature tripping of outdoor lighting due to headlights at night. These do not work well in this application and the Summit model does because it has no delay. We've found that the beam location needs to be adjusted away from the normal helical dust stream in the cyclone to allow the beam to shine across the bin opening.

Since I had the 230V contactor and remote switch for the DC, the sensor costs are only about $50 for all of the parts and enclosures. (see the referenced thread for some sources).

 

[MIKE NOAH]

Don, thanks for the great post!
I have been watching Alan's and Rob's threads on this for some time but have yet to build one. Every time I spend an hour vacuuming and blowing out the filters on the clearvue I swear a sensor is going to the top $#@! to do list but that list changes a dozen times a day it seems. I managed to pick up a couple of garage door sensors from a friend who had to replace their opener but they don't do too much good just sitting on top on the collection barrel waiting on me:no:

 

[Touchwood]

Don,

Your system is intriguing and looks like it would be effective to let you know when the sensors are blocked and it is time to empty the collection bag or bin. But why not make it automatic and have it shut the DC down when the alarm condition is sensed? Alan in Little Washington designed a system for his DC several years ago and I did an adaptation that uses a stock Dusk-to-Dawn photocell (~$12 at Lowe's) that triggers an inexpensive Allen-Bradley "Ice Cube" relay and controls the power relay to the DC and triggers an alarm light as well. I documented it in a thread here more than a year ago and built a prototype for my ClearVue Cyclone. I updated the circuit with a minor revision to provide a service switch (My DC is below my shop on a pad in a tall dry crawl space and I need to go around to an outside door for access to it from my shop). Here's the revised circuit (I now provide power from the remote switch to the candelabra lamp and photocell to simplify the system, but haven't updated the circuit diagram).

In the first picture, 7W Candelabra at top will sit one side of Bin mouth and shines on dusk-to-dawn photoeye in lower handy box. Photoeye stays ''off'' as long as the beam shine through to opposite side. Switch on top handy box controls power to bin light and sensor. This picture, (flash in darkened room) shows unit in normal operating mode.

When the light beam to the photoeye is blocked by sawdust in a full bin (here it is done artificially), the photoeye switches power to its Red output and energizes the ''Ice Cube'' relay coil, tripping the normally open contacts to the closed position and opening the contacts carrying power to the 230 V Contactor. The DC turns off and an indicator light is turned on.

One important note, the photocell sensor selection is important. Many (including most at Lowe's on the electrical component aisle) have a built-in delay of 15-25 seconds to avoid premature tripping of outdoor lighting due to headlights at night. These do not work well in this application and the Summit model does because it has no delay. We've found that the beam location needs to be adjusted away from the normal helical dust stream in the cyclone to allow the beam to shine across the bin opening.

Since I had the 230V contactor and remote switch for the DC, the sensor costs are only about $50 for all of the parts and enclosures. (see the referenced thread for some sources).

Hi Rob,


Thanks for taking the time to read my post and your comments on it. I joined NCWW about 14 months before you posted your approach. I wasn't very active at first so totally missed it..and Alan's as well.


I considered the simplistic “relay logic”approach, but rejected it for several reasons. ...and I didn't want to get all technical at the time I did the post.


I don't like the idea of anything too hot to touch like a 110 volt candelabra bulb being anywhere close to the business end of the DC. I'm thinking it's surface temp is somewhere far north of 150 degrees F. I don't know what the flash point is for fine distributed dust particles suspended in the air around the DC, but I don't think I want to be around to find out. Nor do I like bringing 110 volts in control wires up to that point. Call me old fashioned but I think you're asking for trouble. You've obviously been using the system for some time without incident..but the first incident could be very exciting. Take a look at this report and scroll down to the candelabra incidents around page 15:no:


http://www.cpsc.gov/LIBRARY/FOIA/FOIA08/os/lighting.pdf


Another problem is that cadmium sulphide photocells as you know are like light sensitive resistors..their resistance drops by orders of magnitude with light exposure. CdS is not a high purity single crystal semiconductor and degrades with time..so that it becomes less sensitive to light change. This doesn't matter worth a hoot if the dawn to dusk light comes or goes off 5 or 10 minutes earlier or later..but the effect in the dust bin is to sense obstruction at a lower levels as time goes by. That's probably not a big concern either.


My engineering career designing or managing the design of silicon computer chips for Motorola and GE led me in a different direction. The Genie sensors are really pretty ingenious (no pun) little devices. I opened them up and there's a fairly sophisticated controller board in the sensor. The transmitter sends out a pulsed stream of infrared light below the visible spectrum, and the receiver responds to the frequency of those pulses. The amplitude of the pulses can vary all over (like if the cover glass gets dirty) and the receiver doesn't care. They do this to minimize the effects of ambient light masking the beam (sunlight has all the frequencies in the visible spectrum..including the IR the receiver wants to see). Our circuit uses an LM311 voltage comparator to look at the charge built up on a capacitor by the accumulating bit stream..so it's oblivious to ambient light as well. Since both the sender and receiver are dead cold and the only voltage in the control circuitry around the DC is 12 volts, my first two concerns are overcome. The member in Connecticut tried obstructing the beam by adding more and more pieces of bubble wrap between the sender and the plastic connector hose on the DC (to simulate a dirty connector flex hose). It got to be over an inch thick with no problem..don't know what the thickness would have to be to make it quit..


In talking with Scott Smith, we elected not to automatically turn off the DC when the strobe flashes. Scott's concern is in having the DC quit while he's in the middle of a large planing run and having the planer choke up. He would like to have the option of switching it off after the run. However if you read near the bottom my post I mention that the circuit can do the automatic bit as well. The rating on the TYCO ORHW-SH-112H3F relay is 10 Amps at 277 volts. We're only using it at 12 volts and 180 milliamps...so theoretically you could drive the control coil on the DC contactor. I would prefer another ice cube relay for a dollar or so, for more for isolation.


As for cost..the Tyco relay is less than a buck, the LM311 is 39 cents and the rest of the parts are peanuts. If you want to hook up all the Rs and Cs yourself, the total parts cost is well under 5 bucks. Again I would prefer a professionally made board like this

fully populated and tested. If you don't want to spend the 10 bucks for the super-strobe there are cheaper alternatives. I'm sure you could build it the whole system for well under $50 if that's a hang-up.:gar-Bi


By the way, our member in Connecticut has been talking with Oneida engineering folks and they are highly interested in the design as well.:swoon:

I don't know what they would turn around and sell it for..but we could do it here for a lot less. So it's cheap, reliable and safe. What's not to like??:icon_thum


Take care,


Don

 

[McRabbet]

Lots of excellent points, Don. I understand the concern over having the system shut down during a key operations as you suggest, so perhaps it could be set up to trigger the strobe and then have a manual switch in the relay circuit to then shut down the 230V to the DC blower motor. As for the concern for excessive heat, the candelabra bulbs I'm using run cool enough that I can remove one while it is on withour singeing my fingers. The bulb and photocell are not in the dust stream, but look through clear windows in a collar at the top of the bin; I think explosion hazard is nearly nil. To preserve the life of all of the components, 110V power is coming through the remote switch so the photocell, lamp and relays all remain unenergized until the remote is switched on (I need to redraw the schematic!). I'm using one of these Shop Fox RF remote units that I have used for nearly 10 years and it has been solid as a rock (<$40 at Grizzly, but also available at Klingspors WW Shop).

Truthfully, the main things I do not like about my design is that it runs on 110V and I cannot find a 110 V strobe unit. I could reassign the #1 and #3 contacts on the A-B relay to carry one side of 12V power from a wall wart to a shop strobe (I don't need to switch the Neutrals as shown). Always room for improvement...

 

[Touchwood]

Lots of excellent points, Don. I understand the concern over having the system shut down during a key operations as you suggest, so perhaps it could be set up to trigger the strobe and then have a manual switch in the relay circuit to then shut down the 230V to the DC blower motor. As for the concern for excessive heat, the candelabra bulbs I'm using run cool enough that I can remove one while it is on withour singeing my fingers. The bulb and photocell are not in the dust stream, but look through clear windows in a collar at the top of the bin; I think explosion hazard is nearly nil. To preserve the life of all of the components, 110V power is coming through the remote switch so the photocell, lamp and relays all remain unenergized until the remote is switched on (I need to redraw the schematic!). I'm using one of these Shop Fox RF remote units that I have used for nearly 10 years and it has been solid as a rock (<$40 at Grizzly, but also available at Klingspors WW Shop).

Truthfully, the main things I do not like about my design is that it runs on 110V and I cannot find a 110 V strobe unit. I could reassign the #1 and #3 contacts on the A-B relay to carry one side of 12V power from a wall wart to a shop strobe (I don't need to switch the Neutrals as shown). Always room for improvement...

Rob,

Sounds ike a plan.

Here's the link to the E-Bay strobe light. They come in other pretty colors :gar-Bi
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=110369272314&ssPageName=ADME:B:EF:US:1123

Don

 

[Touchwood]

Don, thanks for the great post!
I have been watching Alan's and Rob's threads on this for some time but have yet to build one. Every time I spend an hour vacuuming and blowing out the filters on the clearvue I swear a sensor is going to the top $#@! to do list but that list changes a dozen times a day it seems. I managed to pick up a couple of garage door sensors from a friend who had to replace their opener but they don't do too much good just sitting on top on the collection barrel waiting on me:no:

Mike,

After you dust off those sensors here's how we mounted them

I can order the board for you if you want, and I must have a dozen 12 volt wall-warts kicking around if you want one.I just posted the link to the EBay strobe to Rob.:icon_thum

Take care,

Don



"It's never impossible, until you walk away "

 

[Travis Porter]

I want one.

 

[JimmyC]

I want one.

Don and I have been waiting for you to weigh in, just pm him.

 

[Travis Porter]

I like this set up better than other scenarios/setups I have seen. It is in the barrell so I don't have to worry about chips getting up the chute, and it doesn't have a motor or something that has to run/spin.

 

[JimmyC]

I like this set up better than other scenarios/setups I have seen. It is in the barrell so I don't have to worry about chips getting up the chute, and it doesn't have a motor or something that has to run/spin.

It's a phenomenally simple set up with a potential to be used in alot of different scenarios. Low voltage is a plus, as is the small footprint. Surprisingly he had few people interested, and he's not making any money on it either. Don's just trying to pay it forward.

Here's his other post on the subject :


http://ncwoodworker.net/forums/f29/group-buy-strobe-light-alarm-22714/

 

[Touchwood]

I want one.

Hi Travis

I figured someone else might pop up, besides JimmyC..so I ordered an extra board:gar-Bi

Don.

 

[Travis Porter]

I glanced at this thread way back and never really had a chance to read it as I have been super busy at work and have some very bad health issues with a family member.

I am glad it got brought back up.

 

[Glennbear]

Thank you very much for sharing this project. I do not use a cyclonic dust collector but the research you and the CT gentleman have done on the Genie photocells and circuit board have got my mental wheels turning on another project. Now if I can only resist the temptation to rip the sensors off my garage door ........:gar-La;

 

[Alan in Little Washington]

That looks like a good, workable approach. I have a couple of comments. The 5 second time delay for that circuit is absolutely necessary. That brings up another issue- Is there anyway to verify the circuit is working? If it fails, what happens? Hopefully it will result in an alarm.

I made another version of my optical unit for a magazine article I just finished (not sold yet). This one is a basic version of my original unit, but a little more compact. Here are some attributes:

• As always, uses commonly available parts.
• All parts, including the candelabra socket/cord are Underwriters Laboratory listed.
• Was designed to require minimal electrical skills to construct, and no electronic skills- can you distinguish wire colors and can you wire an electric socket with a screw driver and wire nuts?

Various comments: The CdS sensor in my unit has been trouble free for 3 - 4 (?) years. If it fails then I will get another from Lowes. The 7W candelabra bulb is not much danger at this low wattage. It is used in various XMAS displays. If I ever get bored with the other stuff I am doing, I will look for a cheap 110 VAC LED to replace it. If the bulb burns out, the system alarms, there is NO danger of it NOT working. Rather than hard wiring the candelabra cord and alarm output, I installed a receptacle- one half is always hot for the candelabra, the other half is switched (powered by the sensor). In the SketchUp drawing below it shows a nominal light socket and bulb being used as and alarm, however anything (within current limits) can be plugged in there- light, horn, strobe, relay to control the DC, or a wallwort to power low voltage versions of these.

A couple of comments concerning this and the LED unit. Both need further design and packaging before they could be commercially marketed, even though both appear to be packaged ok, using available parts.

I wish them luck with Witter at Oneida. I tried to peddle a version of my system to them (as well as Grizzly and ClearVue) a couple of years ago. Witter actually responded. He said he was very interested and going to run it by his staff. I never heard anymore.

As to cost. To be competitive commercially, it would likely need to sell for under $50. This is possible for both the light and garage door LED units. Don't compare it to the overpriced Oneida unit- PSI has a motor/paddle dust bin sensor system like the Oneida, but which only costs $80.

If the CT guys are thinking about selling or licensing this to a commercial manufacturer/vendor I suggest they file a Provisional Patent immediately. I suspect it would likely not be able to be fully patented because of "prior art" and the fact they already posted it here- it is now in the "public domain." Doug Robinson could probably say since he is one of our NCWW patent experts.

Here are some pics of the latest packaging of my system (which I will likely sell if anyone is interested once I get done taking some assembly and other pics for the magazine article):