Official Luthiers Forum!

I recently came across a few threads that talked about building a PID controller for controlling the temperature of heating blankets for side bending. I know it is overkill, however, I must confess that the COOL factor is high with this. I am pretty sure that I know how to hook these things up, but I wanted to incorporate a timer in the circuit. I know I could use a mechanical timer or even simpler, a watch with an on-off switch! But in my mind, PID + Digital Timer = COOL x 2! Anyhow, I have been looking at the PID's from AUBER Instruments and was wondering if it makes sense to go the route of PID (SYL-2342) + Digital Timer (ASL-51) or to go with a PID with Ramp/Soak features (SYL-2352P).

I figure that in order to go the PID + Timer route I will need to add an additional SSR and switch into the circuit. The switch would be to bypass the timer for pre-heating the blanket or to switch the heating blanket into the timer circuit (SSR).

If I go the PID with ramp/soak, I can program schedules in and times in. But they would have to be pre-determined.

Any thoughts? Am I missing or overlooking something with either of these set ups?

Thanks,
Scott

Where's the thread?

I don't even own a heat blanket but being an electronic tech, now I want to build a PID. Where can I find the schematics?

_________________
Hutch

Get the heck off the couch and go build a guitar!!!!

Wow, after doing a little research, probably not enough yet, I'm surprised to find them for $38 everywhere but LMI. If can can use one of those 38 dollar PIDs than I will have lost all interest in building one.

_________________
Hutch

Get the heck off the couch and go build a guitar!!!!

I don't think it's necessarily overkill to have a temperature control for the blanket. Using the PID algorithm may be overkill as on/off control keeps it within a few degrees of set point. On ramp/soak, there is a simple ramp rate limit on mine (Partlow 1160, not ramp/soak programmable) which I think is helpful to minimize hot spots from rapid heat input. Initially I thought a programmable unit would be useful, and probably would be once one's process is really nailed down for each wood, but I like being able to just change the setpoint on the fly rather than entering a new program. Digital timers are certainly cooler, but I find mechanical timers easier to use. I use a 60 min GE motorized timer from HD ($12) which has worked great -- no 'turn past' limit like a spring wound timer, and it has a separate switch to turn off anytime. I wouldn't suggest a watch -- pretty much everyone says that's too risky.

_________________
David Malicky

Well I did some more reading and I don't think I will be using the PID with Ramp/Soak. I just need to decide if I want to use the digital timer or a mechanical one. Is there a way to configure the circuit with using only one SSR and the digital timer?

Sure. Use a 2-input AND gate to drive the SSR's LED (using a suitable resistor in series with the LED to limit its ON current to somewhere between 5mA and 10mA, of course). Connect the AND gate's inputs one to the PID's SSR drive output and the other to a 5V logic timer done/notdone signal from the digital timer......simple!

Potential problems include a) finding a digital timer with a 5V logic done/notdone output, b) the logic sense may be wrong for a simple AND gate, and c) finding a PID controller with a 5V power rail that's available externally.

If you know what I'm talking about at this point, go ahead and make it. If you are confused at this point and/or don't know what I'm talking about, then you should give up and/or get someone who does know what they are doing build it for you.

HTH,
Dave F.

_________________
Cambrian Guitars

"There goes Mister Tic-Tac out the back with some bric-brac from the knick-knack rack"

The timer in your first post has a mechanical relay, so you could just pass the controller's output signal through that relay on its way to the SSR.

_________________
David Malicky

Here's mine. As David suggests, I pass the SSR output from the PID through the relay on the timer to get timed temp control. The small switch under the timer selects using the timer or bypassing it (for preheating, for example). Sorry for the lousy phone picture. The setup works really well, it's definitely overkill, but it's cool! Us scientist types likes our hardware!

Thanks Pat and David! Thanks for the picture Pat. When I read Davids post I immediately thought that I could just use a switch to bypass the timer for pre-heating. If I have things right, it is the "Hot" that runs through the SSR, Switch, and Timer Relay, correct? Are you guys running fuses on yours? I was thinking a 3A fuse for the controller/Timer and a 10-15A for the heating blanket.

Also, Pat, what brand timer are you using there?

Scott

Scott - I found it on ebay. Just search for "digital timer" and you should see some like it. You want one that provides a "countdown" mode, but most of the industrial digital timers have lots of different modes, including countdown. My blanket 120VAC line is fused (10A), but the current through the control side of the SSR is pretty low, so it's not fused. And yes, the hot runs in series from the PID, through the selector switch, the timer and finally the SSR. I inserted an LED that shows whenever the blanket is energized (since the PID "On" indicator is overridden by the mode switch and timer). So far, it works really well!

Pat, Why not run the "hot" from PID through the SSR first so that you can control the pre-heat temperature? If I understand your circuit correctly, the switch bypasses the Timer and the SSR? Or does it bypass the timer and straight to the SSR?

Also, is that a Doolin style bender you got there?

I do actually run the PID when the timer is bypassed - just for the reason you state; control of the temperature while preheating. The bypass switch only bypasses the timer.
The bender is most like the Fox style, although I use Doolin's slat design. Instead of a fixed tower, though, the waist press pivots. The waist press has interchangeable cauls for each body profile. I find this works better with a variety of body styles. The slats hook at the tail end like Doolin's design, and then I bend the waist, and then to the neck block end. You can see the cutaway press at the left end, not used on this body form. My slats aren't under tension like Doolin's, at least, not yet. And the sandwich is stainless slat-blanket-stainless slat-wood-spring steel slat, all attached at the tail end. Works like a charm, and it's a lot faster than bending on a pipe, which is what I've done for the last ten years or so.

I see. I guess both our ways accomplish the same thing. Cool. How about a few more pictures of the bender and a close up of the front panel!

Mine runs an older style Omron H3CA solid state timer. Sort of a fusion between an analogue and digital unit as it is programed by push button and has an LCD count down bar indicator. The control box picks up power from the mains and it splits to an SSR via the PID source terminals. The timer takes power from the closed power switch and daisies supply over to the PID controller which switches the SSR to complete the loop and bring the temp to the set point (SP) from readings taken from the thermocouple.

The PID switched output leaves the SSR to supply a standard panel mount plug socket into which you can plug whatever size heat blanket you like. I chose an SSR with enough output to allow 2 heat blankets to run simultaneously via a double adaptor should I desire, I know that John Mayes uses 2 blankets so I may give that a go some time. It would be no problem to set up as only one blanket would require a thermocouple to control switching for the pair and the PID would certainly assist in avoiding any burning. Next to the power output socket is another smaller flush mount panel socket. This one accepts a quick release plug connected to the embedded 'K' type thermocouple I had manufactured 'into' the blanket for a few bucks extra when I ordered it from MEI. The socket is nifty as it allows different size blankets to be controlled from the same box, you just unplug one, say for side bending and plug the next in which may be for heating up bridges or FB's etc.

I think for accuracies sake, and that if anything is the whole point of a PID temp control box, it is important to maintain the dissimilars of the thermocouple all the way to the back to the PID. You 'can' use copper to hook up inside the box if you wish and the PID will still function. But there is a reason why they sell matching plugs, sockets, and hook up wire and it is not just about over servicing. If you change the resistance of the conductor that the PID has been set to read it cannot know you have done that and will assume in the case of a K type thermocouple that the voltage being delivered is coming from an iron and a nickel alloy conductor. The actual variation in resistance between copper wire and true K type conductors between the back of an input socket and the PID would of course be minute over such a short distance. But when calculations are being based upon the variations in current that is measured in milliamps, then maintaining continuity is a valid concern and probably worth the extra effort if the job is worth doing at all.



To operate the system you plug the control box in to the mains and turn on the power switch (not necessary to have but I had a switch so the box got one 'and' it looks cool) and the red indicator light goes on to confirm the unit is now powered up. The orange indicator light also comes on at this stage and does nothing but let you know that the timer unit is also powered up. (not much point to that really as the LCD in the timer already confirms that 'but' the orange and red makes it look like something really important is about to happen..and that's important) You then set the timer via the push button selectors and the set point for the temp on the PID. Placing your tongue over your bottom lip and to the right side of your mouth you then bite down firmly and hit the green start button.

At this stage the timer will come on and begin the count down display by incremental reductions of the LCD bar, the PID also lights up as does the blue indicator light on the facia of the control box confirming that the PID has closed the circuit to the SSR and power is now being supplied to the output socket. Once the SP has been reached, the blue indicator light goes off and begins cycling through on and off as the PID switching opens and closes to maintain the temprature. Once the timer has finished its cycle, power is then shut down to the PID and output from the SSR to the blanket stops. The red reset button is pushed if you wish to interrupt a cycle at any point and it takes the process back to the stage they had been just before you poke out your tongue. This is a good feature if you wish to run a starter cycle without the sides to pre-heat solid forms if you have them. Its also handy to for a double cook where you heat and bend, allow to cool, and then heat again while still in the form.

The PID can display the current temp, the set point, and an alarm set point at which a separate switch can activate a siren or other warning device if you chose to include one in the loop (am seriously considering this, I have a compressor, an old scuba tank a 145psi air switch solenoid and a spare Clarion from an 1960's Mack that would be fantastic polished up and mounted on the apex of the gable roof of the shed overlooking my neighbours house.)


Q: Are these thinks over kill?

A: NO!!! No more so than a Lie Nelsen plane is to the function of a well fettled Stanley

Q: Why include so many lights and switches and stuff that you don't really need.

A: I needed something that would fit with the look of my Tardis control console.



Cheers

Kim

Looking at the Auber Instruments PID's, is there a reason why one should use the SSR model (SYL-2352) as opposed to the one with the Relay built in (SYL-2342)? The relay says it can handle up to 10A at 120V. It seems like most people who have built these used an SSR.

Also, I assume to use the relay on the Timer (ASL-51), I need to run the neutral that feeds the blanket through the "common " connection, or is there a way to feed only the "Hot" wirde through the timers Relay.


Scott

Most PID controllers have low amp relays, so most of us have to use a SSR or separate relay. If that Auber can switch 10A 120V AND your blanket pulls less than that, a separate SSR isn't needed. (SSRs do have longer life than mechanical relays, which is good for mass production. They're also quiet, but the clicking may be helpful to know how it's cycling.)

Common and Neutral are NOT the same thing. I would find an electrician to help with the wiring as the safety issues are just too important. (Common is the return loop for the low voltage DC circuits and is only for them.)

_________________
David Malicky

Hmm... I should take some better pictures of my setup and post a circuit diagram. I use RTDs, though, for a temp sensor. No real reason, but when I was doing semiconductor research (where I learned a lot of my instrumentation skills), RTDs were a lot more accurate than thermocouples, and they can be easily wired (three lead) to compensate for the wiring between the temperature controller and the sensor.

I second the use of plug in connectors for the blanket and temp sensor. It allows a lot of versatility, and I use a small blanket (1" x 6") for bending bindings for F-holes, and I can use the controller system for controlling it by just plugging the components in. I'll get together some better pics of my setup. I took the best elements from several different designs and incorporated them. (I had a professor that said there was only one actually original idea, and everything since is a derivative from it!) They might prove helpful to one of you all at some time. It works really well for the instruments I build. I use it for acoustic guitars, archtops/semihollow electrics, and mandolins.

I'll try to take some pictures over the weekend.

Here's a schematic of my wiring setup:

Pat,

Thanks for positing. I am going to have to sit down and look this over!

Scott