HELP BE WITH EASY ELECTRICAL QUESTION 4

[cryopumps]

I am a mechanical engineer and want to design an automated Bartender. This is just for fun. I need to know how I can control fluid in a pipe. The pipe will be set up verically and be blocked off by a valve or something. Then I need to do a control of somesort to tell the valve to move at a rate to allow a certain amount of fluid to pass and then close again. I checked on Mcmaster Carr and other and found information on solenoid valve etc. My problem is I dont even know where to start. Is this difficult? I need to keep this off the shelf and as cheap as possible

 

[Mstrvb19]

Cryopumps,
The difficulty of this task depends on a few things. First, how accurate and/or stringent your control requirements are. For example, how accurate and how precise (defined differently) must the flow velocity control be, and/or the shut off velocity. Also, if you are comfortable with mathematical manipulation, such as Laplace transforms and S-domain analysis (or using Z-transforms if you use a digital device for the controller), and can derive a behavioral function to describe the function of the solenoid valve mathematically, then this should be able to handle this. If your control requirements are not that stringent, you could do some rough approximations and avoid the math. Try searching on control system theory/feedback control. It's really an art unto itself. If your requirements are really stringent and none of the stuff I mentioned is familiar, you either have a lot of studying to do, or you can find yourself a good EE buddy.

 

[cryopumps]

I am familiar with everything you said. I want to keep it very easy for the prototype and then make it very precise after. Can you give me any hints on where to start

 

[Barry1961]

You could probably make your own peristalic pumps fairly cheap. Maybe in the $40 to $50 dollar range. They are used a lot for dispensing liquids and would easy to get FDA approval due to intrinsic clean design. Since the duty factor on this should be fairly low you should be able to just use a small plastic rectangle that is rounded on the ends instead of a true cam roller.
If you start getting a large number of bottles it may be cheaper to have multiple pump heads and a single drive motor. One arrangement would be to have the bottles on a carousel with the drive motor on a slide at the pour station. I would guess this would start to get cheaper over 8 bottles.
One bit of experience on the drive coupling is to make one of the "teeth" higher than the other to help it engage.

Barry1961

 

[nbucska]

Watch out for cross-contamination !

<nbucska@pcperipherals DOT com> subj: eng-tips
read FAQ240-1032

 

[buzzp]

I dont think you need any pumps since this is already mounted vertical. Maybe you need to pump it into this vertical column, I dont know. A simple electrically operated valve will work fine as long as it is FDA approved (dont want nasty stuff in your beer). You might lose a little beer trying to figure out how long it should be on to fill a glass (or whatever your filling). This should be pretty easy to make with a relay activated the valve for a predetermined time. To make it simple, just check with some timer mfgs after you select the valve and look for an on delay timer (assume push button for drink). This button will activate the valve for the time dialed in on the timer. Not complicated at all. Check Aerospace, SSAC (entrelec) for cheap timers found at your local electrical house.

 

[pennpoint]

Where have you been? This has been done before.
For hard liquer mostly. (beer would be more difficult to do)
The bottles were placed upside down into a rotary carrage the bartender would turn the carrage to the selected Gin or Vodka (my favorites)or what ever, push a button and a premeasured amount would flow into the glass. I suspect that a solenoid opened for a preset time period then automatically closed.
I havn't seen one in years, not all bars had them.
They did'nt take to the designer bottles too well and the customers felt they were getting short pours.
The bartender did'nt have the freedom of giving the better customer the extra splash when he wanted. It takes all the creativity out of bartending. Puts money in the owners pocket, shorts the buyer.

Does'nt mean you've got a bad idea, keep going, maybe you'll do it better.

Good Luck
pennpoint

 

[jraef]

As an ex-bartender and someone who has done this very thing, I can offer you the cheapest solution yet. Surgical tubing, available at most hospital/medical supply stores, is cheap and sterile and impervious to the effects of alcohol. Use normally closed tubing pinch valves to pinch off the tubing (usually sold at the same places), easy to implement and electrically operated. When you want flow, the valve pushes the plunger on the pinch mechanism open, releasing the tubing. When done or when power fails, the spring return pinches off the tubing again. Use gravity feed by putting a pour spout on your bottle, putting the tubing over the spout, and invert the bottle at some elevation over the dispensory. The trick is measurement. buzzp's timer suggestion will be the least expensive, but from experience I can tell you that the viscosity of the various adult beverages makes the timing tricky. You will need a timer for each different beverage or mixer and a lot of patience to set it up the first time. But as long as you don't change brands (I learned that the hard way too) it should be fine. because I'm a propeller-head I eventually went with a small flow meter/switch I got from Omega

http://www.omega.com/Manuals/manualpdf/M1960.pdf

on the very end of the systems right before the common dispensing tube. It is positive displacement so viscosity was irrelevant. I set up the built-in switch to trip at 1 fluid ounce flowing past the sensor, so I got a shot of booze or 1 ounce of mixer per command. It was a bit of a pain for the mixers though and I eventually switched them back to timing since they were somewhat consistant. The only ones that never worked with this were OJ, lime juice or anything with pulp. Too hard to keep it from clogging.

Have fun. My friends all thought it was cool at first, but after a few rounds nobody cared any more.

&quot;Venditori de oleum-vipera non vigere excordis populi&quot;

 

[Skogsgurra]

I get a little bit afraid when I read the first answer to this question. Aren't you overdoing things slightly Mstrvb19?

Remember that engineering is about designing reliable, functional devices in a sound and economical way and hitting the design target. Not shooting way above it.

By the way, my son-in-law built me exactly one of these "dispensers" for my birthday. There are four bottles (gin, vodka, whisky, soda) and four wind-shield 12 V pumps feeding four separate tubings that end in a common nozzle (nbucska warned about cross-contamination) so that the liquids do not mix until they land in the glass.

It is built into an attache case and even has a brass plaque saying "Port-A-Bar. The Original" on it. Very neat and a big success. It is push-button operated, my next project is to install a PIC to do the metering (time controlled) and also have a code lock to prevent unautorised use. My wife says that it could prevent myself from using it overly; unable to key in the code = no more drinks.

 

[Mstrvb19]

Perhaps I took the question "how I can control fluid in a pipe" in a more mathematical way (assuming he had already chosen a device), and thought the question dealt with control theory, rather than control mechanics. I certainly wasn't suggesting that he do all those things (thus the point that he wouldn't have to if his control requirements are not very stringent).

 

[GOTWW]

I think jraef, was close. Use modern IV dispensing technic. Just keep the bedpan close at hand.