Mark 1 controller for RIMS

[edit1now]

Just spent most of Sunday assembling a Mark 1 version of a heating controller for RIMS and so on.
I have enough bits for two channels, and a big box to put them in (OK, not shiny, that would have cost £120 on its own), but I thought I'd lash together one set to see if it works.

This is the lash-up in a rather second-hand plastic box. The Auber PID controller is driving one of their SSRs, and a PT100 probe from Watch-Hill is plugged-in via a splashproof 9-pin D connector. Boy, do I need everything splashproof! Note the use of sophisticated blanking-off material to cover spare holes in the box (2" wide, sticky...)



In this side view you can see my finger holding the lid of the splashproof mains socket open. The test run was to see if the PID would auto-tune for the kettle, and whether the kettle would stay at 50C which would be my first mashing step. Not enough time to rig up the mash tun, but I have a week off coming up from the 3rd.



I'm still waiting to get the element-is-OK detector circuit working. I've built a dual-channel detector on a piece of Veroboard, but it doesn't detect anything at the moment. Something about differential voltages through an op-amp.

[ECR]

Looks very impressive

[edit1now]

I'll work it out when I get home tonight and post it for you. I have a horrible suspicion that the bits to make a 2-channel controller (one channel for RIMS and one for sparge water) will come out at something over £200

[edit1now]

Ignoring the element detector thing, the rest has been pretty straightforward, apart from fighting the crimp terminal pliers

[Andy]

I'll have to take a pic of the insides of my RIMS PID controller box, it's crimp terminal heaven/hell depending upon your point of view

[edit1now]

When I do crimp terminals, they're either absolutely fine first time, or something goes wrong and I have to chop them off and start again. OK when you have a sackload of them, less good when they come in little packs of 10 from two miles away. Then the little wire, which I measured to go from here to there, is a bit shorter, and gets tight, and maybe I should cut a new piece...

[edit1now]

OK here are the costs for the controller as you see it in the pictures:

Auber Instruments:
2 x Universal 1/32 DIN PID Temperature Controller (SYL-1512) = $69.90
(1 x K Type Thermocouple (6 ft. cable) (TC-K6) = $6.50 Not used here)
2 x 25A SSR (RS1A40D25) = $30.00
------------------------------------------------------
Sub-Total: $106.40
Zone Rates (Shipping to GB): $20.00
Total: $126.40 = £63 for 2 sets, and I didn't have to pay duty & VAT for some reason.

Farnell:
1082451 2 10.18 SWITCH, DPST ILLUM RED; Switch function KG412A2XDA26X6 APEM
1170085 2 11.78 PLUG, D IP67/66 FREE 9 WAY; LTWDB-09BMMA-SL7001 LTW
1170070 2 11.78 SOCKET, D IP68/67 PANEL 9 WAY; LTWDB-09PFFS-SL7001 LTW
+VAT but free postage = £38.94 for two sets

Watch Hill
PT100B 3mm x 150mm S/S Housing 2m 3 Core + outer ETFE £16 each + £2 p&p + VAT = £21.15


Wickes
Exterior 2-gang socket £19.99 each
A plastic cable gland, which I can't find on their website: I also have a couple from Farnell but I forgot to buy back nuts for them.
13A plug 63p I think.

Maplin
various crimp terminals, maybe £1.50 worth

Scrap box: lengths of stranded and solid insulated wire, from e.g. 2.5mm T&E. Curver plastic box.

So for one set £31.50 + £19.46 + £21.15 + £19.99 + (guess) 99p + 63p + £1.50 = £95.52 (gulp!)

It would be cheaper to buy an ATC-800 but less challenging, methinks.

The Mark 2 will go as two separate channels in the big metal box mentioned above, with a big red 4-pole isolating switch, plus LEDs for Heating On and Element OK (If I can get it to work), the switch-mode power supply from an old disk drive enclosure (to get 12 volts for LEDs etc.), all fed from two different 13A plugs, so I can plug in to different rings in my house (it's OK, I've only got one phase!).

The big illuminated splashproof rocker switch doesn't illuminate:(

I think you can get cheaper three-pin splashproof sockets, though the boaty ones from Force 4 are about the same price as the D-plugs from Farnell.

[Graham]

I'm still waiting to get the element-is-OK detector circuit working. I've built a dual-channel detector on a piece of Veroboard, but it doesn't detect anything at the moment. Something about differential voltages through an op-amp.

What you are trying to do is quite difficult, well, not trivial, due to the difficulty of maintaining galvanic isolation. Presumably you are using a current transformer, something like this:
link to RS current xformer Which would make life fairly easy if you are.

You also have to take account of the fact that a decent PID, for heating, pulses at about a one-second burst rate. You have to make sure that it doesn't trigger your alarm during the off period, which is best done by gating it with the firing pulse, although a long time-delay will do as a second best.

Pity really, because the software in the controller could have done this quite painlessly without any additional hardware, had they thought of it.

[edit1now]

Thanks Graham. I have a current transformer from Farnell, which gives about 2.5V peak-to-peak from a 2.2kW kettle element. I'm just hoping to get my friend, in the Engineering department at work, to look at the differential voltages in the circuit. All I'm after is an LED to come on, even if it pulses on and off, to tell me I haven't burnt-out the element.

[Andy]

At the risk of stating the obvious but is it worth it ? What situation are you trying to detect and cope with ? Seems that you're wasting valuable beer brewing time/cash

[Aleman]

At the risk of stating the obvious but is it worth it ? What situation are you trying to detect and cope with ? Seems that you're wasting valuable beer brewing time/cash

But then you would know that the element is f*cked, and not try to continue to use it in your RIMS setup for 4 batches

I'm sampling the Coopers tonight to see where the gravity is, and then its a case of planning the CAP Brewday to coincide with good weather next week. At least this time with a new element I don't have to worry about the temperature falling to 60C or lower during the mash

[Andy]

But then you would know that the element is f*cked, and not try to continue to use it in your RIMS setup for 4 batches

I could work that out quite easily by looking at the PID temperature display.

[edit1now]

I think that when I was looking at the whole posh controller thing, the additional £15 for the current transformer and detector circuit seemed to be worth it. It won't be the end of the world if I have to leave it out, but if it does work then it will stop me waiting two hours for a boil which doesn't, after a mash that was OK until the element got sticky stuff on and threw its (one-way) trip.

[edit1now]

Yes, I definitely need the little light to say the element's OK

[edit1now]

The splash-proof illuminated switch (Farnell part number 1082451), mentioned above, doesn't light up because the illumination is a 12V LED. Their website is not very clear on the topic: I had to read the manufacturer's PDF and work it out from the part number. The switch looks as if it's meant for a lorry dashboard or something, but will switch 20A resistive at 240V. I must have blown the LED the first time I tried it.