solenoid valve driver

[paulrevelcet]

Debating over ssr's or transistors to drive some solenoid valves. (there are 48 of them)<p>Here is the run down.<p>My control is a digital i/0 board that can sink .064 amps or source .015 amps at 5 volts.<p>The valves I have are 24 volts they draw .67 watts or .279amps. <p>now I hate to spend 5 or 10 dollars on 48 solid state relays when I can use transistors to switch the 24 volt power and due to the very small amp requirement ssr's might me overkill.<p>One more question I am using a i/o board from cyber research inc. part# cydio 48hp (it goes in a pc's pci slot) and even though it doesn't say I was just wondering if I need to use opto isolators on the outputs of the card. Thanks.

[dyarker]

Is that 24 Volts AC or DC?<p>Probably AC, and by the time you buy parts it would be around $4 per solenoid. DC a little less.<p>Yes you need opto-couplers.<p>I can draw a circuit, but need to know if AC or DC.

[paulrevelcet]

There dc solenoids, made by 3m part# EVO-3M-24
very low power. Also I think im going crazy because I cant seem to figure out the right transistor to use in relation to the hfe min.<p>I mean if I only have 10ma coming out of each output then I need a transistor that will switch
24volts, 27.9ma using 5 volts at 5 to 10ma. This is all dc voltage so not a very demanding job but I cant seem to get my mind around it, and I guess I will have to use diodes across the solenoids, its late and I think after I sleep on it I will forget all about this project and that will solve everything. naw, probably just have nightmares about being back in school and not being able to remember a right triangle from my hypotenuse. <p>thanks for the help.

[dyarker]

Its morning here, I'll take a stab at it. Sleep easy.

[Edd]

Your I/O board certainly has the requirements for bipolar driving. The optical isolator use would lean towards the safety of power isolation on the interfacing.
If you were worrying on the HFE on the bipolars, I have used 2 discrete bipolars hooked into a darlington configuration where I custom tailored the pairs effective gain to a precise HFE by the changing of 1 ..(or 2) resistor values in their intercoupling. For example see this ref data sheet:
http://www.nteinc.com/specs/2300to2399/pdf/nte2341.pdf
Page two
Notice the biasing resistor at the bottom left schemas , in manufacture these are laser trimmed at the die test stage of fabrication to make units fall within the devices HFE spread spec. Sometimes even a second biasing resistor is used between the first transistors base and its emitter so that a balanced/sharing of gain can acquired for each stage.
And yes, a spike damping diode across each solenoid.
BTW what is the end use/function of sooo MANY little valves ?<p>73's de Edd
[email protected] .........(Interstellar~~~~Warp~~~Speed)
[email protected].........(Firewalled-Spam*Cookies*Crumbs)

[Chris Smith]

SCRs for DC and Triacs For AC, cheapest and most bullet proof way to go. All industrial systems almost exclusively use them.

[paulrevelcet]

Well Mr.Whatley, The place where I do some parttime work, a skating rink and fun center with games and such, fell into some old audio animatronics from chuck e cheese I think one of the characters is called elvis, he is a big ol bear with a snarled lip, any way, Im trying to help them convert these things over from a reel to reel tape player that controls them to computer control which has led to a digital i/o card and some software called light o rama that allows you to sinc music to on off signals, the hope is we will be able to rebuild some of these into something that resembles the original, and not spend alot of money, I am a iscet (certified electronics technician ) but due to the location I live in, I work full time for the chamber of commerce as a building supervisor because I cant find work in electronics, so I get a little rusty between projects.

[rshayes]

I assume that these valves are controlling compressed air. In that case, the solenoids do not have to be referenced to any particular ground. If the 24 volt supply only supplies solenoids, it can be grounded to the same ground as your interface card, and isolation should not be necessary.<p>Bipolar transistors should be adequate for the job and will probably be cheapest. I would look for an NPN transistor in a TO-220 case with a 60 volt or above collector voltage rating. This should be rugged enough for the solenoids you describe. The metal tab can probably dissipate enough power without an additional heatsink. Smaller transistors, such as TO-5 and TO-92 cases, are marginal. Cost is probably around 70 cents each in small quantities for the TO-220 devices.<p>Transient supression will be needed. Usually, this is a diode across the solenoid that conducts when the solenoid is switched off. This results in a slow release characteristic. The turn off time can be reduced by allowing the solenoid voltage to rise above the 24 volt power supply, and limiting it with a zener diode. If 48 zener diodes are too expensive, a smaller number of zeners can be shared among several solenoids by using regular diodes, which are dirt cheap.<p>When transistors are used for switching, a beta of about ten is usually assumed for design purposes. Most transistors have betas well above this so saturation is guarenteed. Your interface board can source 15 milliamps, but I would design for about 5 milliamps of base current. This should insure saturation with a collector current of 30 milliamps. Base voltage will be about .7 volts, so the base resistor should be less than 860 ohms. Either 680, 750, or 820 ohms would work.<p>If you need more drive current, an emitter follower using a small signal transistor can be added between the interface card and the solenoid drive transistor. This would also reduce the chance of transients generated by the solenoids getting into the interface card.<p>Parts cost per solenoid should be under a dollar at the worst.

[amuron]

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by Chris Smith:
SCRs for DC and Triacs For AC, cheapest and most bullet proof way to go. All industrial systems almost exclusively use them.<hr></blockquote><p>I beg to differ on that one. Forklift controllers used to use SCR's with autocommutation circuits for DC control, but I think many have changed to IGBTs. However I could be wrong.<p>As far as AC control, I know of no industrial units that use triacs. Every power controller i've ever run into uses SCR's. Per memory its due to dV/dT issues with triacs, to say nothing of power handling. We used to build a wide range of industrial power controllers, 30A-240V all the way up to 3 phase 480V 600A controllers, and they all used SCR's.<p>Ron

[Edd]

Plus, if using SCR’s in the prescribed application, seems like repetitive activations of the valves will be in order to enact the mechanics. There would be further circuitry complexity involved in enacting a speedy unlatching of SCR’s for the next quick response.
One possible option, being the utilization of positive nodes of rectified AC as the SCR’s power source, if the resultant electromechanical/ acoustical hum level at the solenoid did not present a problem.<p>Aside: Paul:
Assuming that assigned multiple utilizations of remote slave cylinders potentially might be involved in performing a single mechanical action. Have you been able to ascertain how many channels of servo control (different tone frequencies on the original tape) are involved in the control system ? Also the control PCB electronics is probably of early eighties vintage …right ? (A polling of several date codes on IC’s on the board should guesstimate that.)<p>
73's de Edd
[email protected] .........(Interstellar~~~~Warp~~~Speed)
[email protected].........(Firewalled-Spam*Cookies*Crumbs)

[paulrevelcet]

Yes Mr.Whatley, The Tech. is 80s but I cant tell about how many channels of servo control they used,the equipment is non functional and in about a gagillion pieces. Were still going through the PILE. (which I might add is a very appropriate word for the twisted and tangled mess of body parts, come to think of it, it looks like an explosion at a mascot convintion) I think I will give the transistors a go. If we build something worthy of attention I will put it up on my web space, that is if any one thinks they might be interested. For now thanks for all the advice, I could gush on for days about how great I think it is that people have a place to go to get help with there questions, but I wont, only to say that, well, its great.