Electronic Cigarette Smoking Machine Controller Needed

[ronv]

Finding a simple a off the shelf turn key controller isn't looking all that simple. I think I'll send the guy who built my controller a link to this forum and see if his electronics guy can come up with a better box. I want one where it is easy to set a larger number of puff duration and frequency times than the one I have now. I would also like to be able to set the number of puffs and then stop. Also, a nice digital display showing the settings and with timer displays would be great.

If you want to see what commercial t-cigarette machine look like check out borgwaldt: https://borgwaldt.hauni.com/en/instruments/smoking-machines/linear.html. I think they start at $30K.

Thanks for all your comments and help.

I don't think it would be difficult to do in hardware. The biggest cost would be in the printed circuit board, but if other people want some that cost would be reduced as you probably can buy 5 for the same price as one. Maybe you could put the specs you would like in one post and I'll have a go at a hardware design. Then we can figure out how to get it built.

 

[Chris Havel]

I don't think it would be difficult to do in hardware. The biggest cost would be in the printed circuit board, but if other people want some that cost would be reduced as you probably can buy 5 for the same price as one. Maybe you could put the specs you would like in one post and I'll have a go at a hardware design. Then we can figure out how to get it built.

You help is very much appreciated. I purchased most of the parts for the machine from Teague Industries in Woodland California. They should be able to build the design.

We are trying to make a machine that emulates e-cigarette users. From clinical and other studies we see that the average user takes a 4 second puff every 30 seconds. It would be nice if we could control the machine to take a 3.5 second puff every 40 seconds. It just seems like a good idea to be able to be flexible in programming the machine.

For the puff duration we would like to supply 12 volts to both the solenoid and the 3-way valve for puff duration times ranging from 0.5 to 10 seconds. We would like have a puff frequencies ranging from 20 to 180 seconds. It would be great to have a puff counter so the puffing would stop after a predetermined number of puff say 3 to 60 puffs.

It would be great if we could control the solenoid which directs the power from the power supply and the valve which directs the suction from the vacuum pump independently. E-cigarette users can push the button on the battery for a second or 2 to preheat the coil before puffing. Right now I control the solenoid which closes the circuit to deliver the power to the coil and open the 3-way valve to the e-cigarette at the same time using the same 12 volts from the controller. A controller that would allow for the power to be delivered to the coil for a 0.5 to 2 seconds before the second 12 volts was sent to the valve would allow for us to simulate this.

The main reason why I want to have an smoking machine is so I can simulate the e-cigarette smoker and collect the vapor in the gas washing bottles for chemical analysis. The biggest driver in e-liquid vaporization and delivery is the voltage. The amount of e-liquid vaporized by a device can go up 8 fold from 3 to 6 volts. Also at higher voltages a lot more formaldehyde is produced. I have very good control over the power (voltage and amperage) being delivered using the constant voltage power supply.

Here is a link to a U-Tube video that shows the machine: running

 

[ronv]

You help is very much appreciated. I purchased most of the parts for the machine from Teague Industries in Woodland California. They should be able to build the design.

We are trying to make a machine that emulates e-cigarette users. From clinical and other studies we see that the average user takes a 4 second puff every 30 seconds. It would be nice if we could control the machine to take a 3.5 second puff every 40 seconds. It just seems like a good idea to be able to be flexible in programming the machine.

For the puff duration we would like to supply 12 volts to both the solenoid and the 3-way valve for puff duration times ranging from 0.5 to 10 seconds. We would like have a puff frequencies ranging from 20 to 180 seconds. It would be great to have a puff counter so the puffing would stop after a predetermined number of puff say 3 to 60 puffs.

It would be great if we could control the solenoid which directs the power from the power supply and the valve which directs the suction from the vacuum pump independently. E-cigarette users can push the button on the battery for a second or 2 to preheat the coil before puffing. Right now I control the solenoid which closes the circuit to deliver the power to the coil and open the 3-way valve to the e-cigarette at the same time using the same 12 volts from the controller. A controller that would allow for the power to be delivered to the coil for a 0.5 to 2 seconds before the second 12 volts was sent to the valve would allow for us to simulate this.

The main reason why I want to have an smoking machine is so I can simulate the e-cigarette smoker and collect the vapor in the gas washing bottles for chemical analysis. The biggest driver in e-liquid vaporization and delivery is the voltage. The amount of e-liquid vaporized by a device can go up 8 fold from 3 to 6 volts. Also at higher voltages a lot more formaldehyde is produced. I have very good control over the power (voltage and amperage) being delivered using the constant voltage power supply.

Here is a link to a U-Tube video that shows the machine: running

I think I almost have it.

Let me play it back to you and see if we are on the same page.

You would like a puff duration of 1 to 5 seconds.
You would like one puff cycle to be adjustable from 20 to 40 seconds.
You would like to be able to delay the vacuum valve from .2 to 2 seconds after the coil voltage is applied.
You would like to be able to set the number of puffs from 3 to 60 and you would restart it after each of these?

Now a couple more questions:
What do you mean by a frequency of 20 to 180 seconds?
How would you like to select the times? I envision 2 rotary switches. Say 5 positions for duration - 1 to 5 seconds. One for cycle time - again maybe 5 positions, 20, 25, 30 35 and 40 seconds. And finally maybe a dip switch - a small switch with several positions to set the number of cycles.
We could also make them continuously variable but then you would need something to measure them with.
How accurate do the times need to be? Could 1 second be 1.1 seconds and 5 seconds be 5.5 seconds? Or better?

 

[KeepItSimpleStupid]

I think what I said earlier was valid. You need a much better HMI interface. I do think the Arduino would be a decent platform with the IDE being the HMI.

Some disadvantages, yes. Not 100% stand-alone.

==

I was going to mention "power control" from an electronic load rather than power supply, but your simulating what actually happens.

My "all scenario" brain says the battery is constantly loosing charge and does the model account for that or does it have to?

 

[Chris Havel]

I think I almost have it.

Let me play it back to you and see if we are on the same page.

You would like a puff duration of 1 to 5 seconds.
You would like one puff cycle to be adjustable from 20 to 40 seconds.
You would like to be able to delay the vacuum valve from .2 to 2 seconds after the coil voltage is applied.
You would like to be able to set the number of puffs from 3 to 60 and you would restart it after each of these?

Now a couple more questions:
What do you mean by a frequency of 20 to 180 seconds?
How would you like to select the times? I envision 2 rotary switches. Say 5 positions for duration - 1 to 5 seconds. One for cycle time - again maybe 5 positions, 20, 25, 30 35 and 40 seconds. And finally maybe a dip switch - a small switch with several positions to set the number of cycles.
We could also make them continuously variable but then you would need something to measure them with.
How accurate do the times need to be? Could 1 second be 1.1 seconds and 5 seconds be 5.5 seconds? Or better?

We're almost there:

Puff frequency and puff cycle are the same. I would like the cycle to be variable from 20 to 180 seconds to simulate puffing from 3 times per minute to once every 3 minutes. 20, 30, 60 90 and 120 would work.
I would like to be able to adjust the puff duration from 1 to 10 seconds with 0.5 second or better accuracy. Times of 2, 3, 4, 5, 6, and 10 second duration are much better than I have now.
I would like to be able to adjust the puff duration independently for the valve and the solenoid. This way I could have a power duration of 5 seconds and a vacuum for 4. This would mimic a user activating the battery for a second before he takes a puff.

Rotary switches are good with fixed times and 10% accuracy will work and would be better than variable times for reproducibility.

We're seeing a coefficient of variation of 18% in the delivery/vaporization of e-liquid from 15 puffs from the same device. Delivery is very easy and accurate to measure with e-cigarettes. We simply weigh the device on an analytical balance before and after puffing. Typically we get 80 mg delivered in 15 puffs at 3.7 volts.

If you can design a controller that can do this I'll put you on the paper as an author. You will have the gratitude of the everyone doing e-cigarette research and the FDA.

 

[KeepItSimpleStupid]

FWIW: LabView would be cake.

 

[gary350]

This is what you need. Dual timer relay 0-180 seconds ON and 0-180 OFF. Use the relay to turn ON/OFF your 12 volts.

https://www.ebay.com/itm/AMF-POTTER...658?pt=LH_DefaultDomain_0&hash=item2a4d7ecb5a

I think your data is going to be totally different than a real e-cigerette. You need to test each brand of e-cigerette. I think it will be more important to know what each brand of e-cigerette does than your smoking machine.

 

[Chris Havel]

I think your data is going to be totally different than a real e-cigerette. You need to test each brand of e-cigarette. I think it will be more important to know what each brand of e-cigerette does than your smoking machine.

We are testing real e-cigarettes. We bought close to a hundred KangerTech MiniProTank 3 models for testing and for a clinical trial. We are vaping all kinds of commercial e-liquids with them. The number of different devices and e-liquids is daunting. Thousands of combinations not to mention different voltage batteries. I won't even go into Rebuildable Atomizers where people roll their own coils to the resistances to their liking. Google sub ohm vaping. At least it's teaching Ohms law to a lot of people.

Thanks for the link. I'll check it out and consider buying one.

 

[Chris Havel]

FWIW: LabView would be cake.

Labview is definitely on the radar. We have a site license for it.

 

[Dr_Doggy]

the nice thing about micro-controller is that all you need is ( 5v trigger-12v capacity relay)/resistor/transistor/diode to interface with device, and then with simple programming you can set any on./off times you like,but then you need to think of pcb or breadboard, and minor circuitries plus programming.

this is the basic connection you need to attach microcontroller to each thing you want to control, you could use one relay for main control for all , or a relay for each 12v line:
**broken link removed**

and with more complex programming and peripherals(ie keyboard, buttons, data terminal, bluetooth socket) you can setup so that all you need to do is input settings instead of needing to rewrite code each time, but then you would need to consider a input device to run each setting.

Personally, 4u, i still lean towards the controller in post 3, all you need to verify is that this device is actually able to support your resolutions. after that hardware is done, this device looks easy to program with "flow code". I presume "flow code" used would be similar to this compiling program(but didnt verify) which is what i used when i first didn't know micro controller programming as well:
http://gcbasic.sourceforge.net/aboutgcgb.html
in that example the relay was put on port b0, and sends a 1second pulse then waits a second.

One project i did was i took an ATX power supply, which supplies lots of current at both 12v and 5v, with that my microcontroller supplied by the 5v was able to connect direct to the pin that turns the power supply on and off, so no extra parts, I then attached this bluetooth device(http://i00.i.aliimg.com/wsphoto/v0/...uetooth-b-font-font-b-RS232-b-font-serial.jpg) however this may not be ideal for you if your power supplies are variable. totalling 2 parts. and a bit of code, again, it may get tricky processing terminal commands for you,
so if you skipped the bluetooth, you could just cut a few wires from the ATX supply and solder it direct to the micro controller bed

EDIT: sry didnt see other threads, if shoe fits Labview may be way to go

 

[Chris Havel]

I think what I said earlier was valid. You need a much better HMI interface. I do think the Arduino would be a decent platform with the IDE being the HMI.

Some disadvantages, yes. Not 100% stand-alone.

==

I was going to mention "power control" from an electronic load rather than power supply, but your simulating what actually happens.

My "all scenario" brain says the battery is constantly loosing charge and does the model account for that or does it have to?

I know I couldn't stand the unknowns with batteries. My cheap PS loses a bit of voltage under load so I adjust it with a VOM. I have a button pushing solenoid for the battery that hooks up instead of the solenoid so I can compare delivery witjh 3.7 V batterie

 

[ronv]

="Chris Havel, post: 1219825, member: 257028"]

We're almost there:

Puff frequency and puff cycle are the same. I would like the cycle to be variable from 20 to 180 seconds to simulate puffing from 3 times per minute to once every 3 minutes. 20, 30, 60 90 and 120 would work.
I would like to be able to adjust the puff duration from 1 to 10 seconds with 0.5 second or better accuracy. Times of 2, 3, 4, 5, 6, and 10 second duration are much better than I have now.
I would like to be able to adjust the puff duration independently for the valve and the solenoid. This way I could have a power duration of 5 seconds and a vacuum for 4. This would mimic a user activating the battery for a second before he takes a puff.

How about this:
Can we forget about total time? Then we could use valve on time for the totals. If we do it that way the delay from "burn" to puff will fall out and the total time will vary depending on the delay plus the puff time.
For example: Lets say we set the time between burn and puff to be 0.1 seconds and puff time of 2 seconds with a duration of 3 seconds, with a puff count of 5. The total time would be (.1 + 2 +1) X 5 or 15.5 seconds. Then if we change only the delay between burn and puff to .2 seconds the total time would be 16 seconds.

Rotary switches are good with fixed times and 10% accuracy will work and would be better than variable times for reproducibility.

We're seeing a coefficient of variation of 18% in the delivery/vaporization of e-liquid from 15 puffs from the same device. Delivery is very easy and accurate to measure with e-cigarettes. We simply weigh the device on an analytical balance before and after puffing. Typically we get 80 mg delivered in 15 puffs at 3.7 volts.

Lets shoot for better accuracy then. I can make the clock for the timers adjustable so they can be set more accurately.

If you can design a controller that can do this I'll put you on the paper as an author. You will have the gratitude of the everyone doing e-cigarette research and the FDA.

That would be nice. Maybe you could let me build it for 75% of what the other guys quote you.

 

[Chris Havel]

Lets shoot for better accuracy then. I can make the clock for the timers adjustable so they can be set more accurately.

I think the high variability is due to variation in the e-liquid permeating into the cotton wick surrounding the coil. Longer puff intervals might clean it up.

That would be nice. Maybe you could let me build it for 75% of what the other guys quote you

Let's get a price from them. I think we should build at least 5. I know of one lab that wants one already. You could sell them on ebay.

 

[KeepItSimpleStupid]

Chris:

You can use constructs like:

[quote=someone]Something you wish to quote[/quote]

which will turn out as:

Something you wish to quote

"More options" will show a "preview" tab.

 

[ronv]

OK, lets try one more spec. I have one more question:
On the puff counter from 3 to 60..... It is hard to find a switch with much more than 10 positions. There are a few options:
We could count by 3's from 3 to 30 or perhaps 4 to 40 or 5 to 50.
We could use individual switches and count 1, 2, 4, 8, 16, 32 and you could add them in your head. My favorite would be the first option.
So here is the sequence as I see it.

Reset
Start
Turn on coil
Delay .2 to 2 seconds
Turn on valve
Time from 1 to 10 seconds
Turn off valve and coil.
Count as one cycle
Check if counter matches setting
If not start another cycle.
If yes - stop

Can you post a link to your relay and valve so I can see the voltage and current?
We could build the relay into the box if you like.

 

[Chris Havel]

The timings look great except would it be possible to include a 0 second delay on the coil?
I'm not sure what you mean by the first option?
I don't have links to the relay and valve but both are 12 volts. I think the relay can handle 10 amps.
Both devices are being driven by the same controller which is fed off a little wall wart. I can put my meter on the connections and see what it says.
I got the relay at Electronics Plus for $24. Having it built in would make it more marketable, but for now, we can leave it external.
When I get to work tomorrow, I'll take some more pictures provide you with the information.

 

[ronv]

="Chris Havel, post: 1219895, member: 257028"]

The timings look great except would it be possible to include a 0 second delay on the coil?

How about 0 to 1.8 seconds in steps of .2 seconds?

I'm not sure what you mean by the first option?

Since it is difficult to find a rotary switch with more than 10 positions I would suggest making the puff counter count by fives instead of ones. So you could dial in 5 puffs to 50 puffs in increments of 5..... 5, 10, 15, etc.

I don't have links to the relay and valve but both are 12 volts. I think the relay can handle 10 amps.
Both devices are being driven by the same controller which is fed off a little wall wart. I can put my meter on the connections and see what it says.
I got the relay at Electronics Plus for $24. Having it built in would make it more marketable, but for now, we can leave it external.
When I get to work tomorrow, I'll take some more pictures provide you with the information.

I just need to know the current so I know how big the transistor to turn them on needs to be. I think you said they were both 12 volts.

 

[ronv]

Oh, I see now it is not a relay, but a solenoid to push the button.

 

[Chris Havel]

Oh, I see now it is not a relay, but a solenoid to push the button.

Here are some better pictures which might help you see what's going on. I was wrong about the 12 volts. It's 15 volts and 15.8 on the meter when the controller fires.
The battery solenoid picture in the above post is what I use when I want to test EGo type batteries for comparison to the power supply. It is replaces the relay switch for the power supply when testing battery power.
Let me know any other information that you need.

 

[KeepItSimpleStupid]

Just a stupid question since I have no interest in e-cigs.

They basically just have a power switch? I suppose you decouple or want to de-couple the power from the draw? i.e. Make them separate.

e.g.
1) Force the switch closed (i.e. tape) and apply power some other way.
2) Modify the e-cig, so the power switch is used.

The e-cig vents when the solenoid is off and goes to your apparatus when engaged.

There should be no reason to substitute anything but a constant voltage power supply for the e-cig supply?

==

FWIW: Then there are these controllers: **broken link removed**