Advice on solenoid valves and PIDs

[pjbiker]

Hi all. As I've just finished my latest upgrade to all SS, and I have a pumped system, I'm thinking the next natural progression (after sight tubes) is to HERMS, with 2x PIDs. I'd use a coil of 8/10mm copper in the HLT as the heat source with temp probes on HLT temp and Mash tun inlet. As my pump is 12V anyway, I'd really like to keep the rest of it 12Volt. I can see 12V PIDs for about £30. The 12V solenoid valves seem hard to find in any variety but I have seen some cheap (virtual village) £6 each but only available in 'normally closed'. I plan to leave the pump running but diverting to and from heating coil depending on mash tun inlet temperature using a pair of valves.

My question is, is it possible to set a PID to simultaneously open one and close the other valve if they are both 'normally closed'?

 

[drut]

With 12v it is a doddle - you need a resistor, a transistor and a diode.
Alternatively you could use a 12v relay with no/nc contacts.
Can you do this yourself, or do you know anyone who can?

You will need to know the way the PID is capable of driving the solenoids e.g. open collector/drain, output (sink?) current etc.

If the PID is capable of both sourcing and sinking enough current you would just connect one wire each of the two solenoid valves together, then the other side of one goes to +12v, the other side of the second goes to 0v. (Don't forget to add diodes to pass the back emf spike!).

If this is all gobbledegook, you need to find someone ( :geek: ) local to you!

 

[Springer]

I am not sure what your plan is with the solenoids PJ, you to lazy to turn a valve, or are you going for a sit back, watch and do nothing system:lol:
With HERMS you will need a 240 volt heat source, so why not control it with a pid and a power relay, form Virtual Village or wherever.

 

[pjbiker]

With 12v it is a doddle - you need a resistor, a transistor and a diode.
Alternatively you could use a 12v relay with no/nc contacts.
Can you do this yourself, or do you know anyone who can?

You will need to know the way the PID is capable of driving the solenoids e.g. open collector/drain, output (sink?) current etc.

If the PID is capable of both sourcing and sinking enough current you would just connect one wire each of the two solenoid valves together, then the other side of one goes to +12v, the other side of the second goes to 0v. (Don't forget to add diodes to pass the back emf spike!).

If this is all gobbledegook, you need to find someone ( :geek: ) local to you!

Simple answer is 'yes' it can then. Except it's not so simple. Sure I can work it out with a :geek: and a Maplin's catalogue when the time comes. I think I understand.

 

[pjbiker]

I am not sure what your plan is with the solenoids PJ, you to lazy to turn a valve, or are you going for a sit back, watch and do nothing system:lol:

Precisely, I want to leave the PID to maintain mash temperature for the 90 minutes of my mash. Isn't that what they're for?

With HERMS you will need a 240 volt heat source, so why not control it with a pid and a power relay, form Virtual Village or wherever.

Yes, I guess I'll need a 240V relay to control the element on the HLT but the PID that controls that relay can be a 12V one can't it?

 

[pjbiker]

Looking at these

http://cgi.ebay.co.uk/Electric-Sole...ial_Automation_Control_ET&hash=item230cff896f
http://cgi.ebay.co.uk/12V-24V-32V-P...716?pt=LH_DefaultDomain_0&hash=item43a485f6c4

still can't work out the 'clicky' thing on the forum.

 

[drut]

It makes perfectly good sense to me to use the HTL as the heat source, it does not even have to be accurate or constant because the PID feedback loop will compensate.
If you use one of your PIDs with a SSR to control the HLT temperature that will be fine (actually overkill as you only need a straightforward temperature controller, (eBay about £19 inc carriage) as long as your element takes 10A or less.

I use several TET612 controllers, but mine are 240V and I use the SSR outputs.
It looks from the diagram that J1 is a relay with no and nc contacts. These should drive your two solenoid valves with no additional parts.

I would suggest using a PT100 probe. This will give greater inherent precision and the TET612 will read to 0.1C with it. When using a thermocouple, the TET612 will read to the nearest 1C and you have to be much more careful with calibration.

The downside will be the somewhat "lumpy" control. Being bypassed or going through the HLT coil will be very coarse and might benefit from a high flow rate to ensure that the returning liquor/sweet wort does not gain very much heat - this would result in de-naturing the enzymes and stopping conversion.
I personally use RIMS and raise the mash to 77C for mash out - this would be feasible for you if your HLT is at 82C or so.

You might have problems with the hydrodynamics - bang on/off of the two valves might shake things about a bit and possibly upset the pump.
I can see other possible problems with time constants but it would be a question of suck it and see - Good Luck!

 

[Springer]

Drut has the right words Solid state Relay, ssr, thats what I was thinking :? Mine was from Virtual Village, we "members" have/had a 20 % discount code for use there
The idea is to keep the pump running to aid temp stabalilation and filtering, all the time but have the pid control the heat exchanger / wort temp via a ssr. :?
S

 

[BarnsleyBrewer]

It makes perfectly good sense to me to use the HTL as the heat source, it does not even have to be accurate or constant because the PID feedback loop will compensate.
If you use one of your PIDs with a SSR to control the HLT temperature that will be fine (actually overkill as you only need a straightforward temperature controller, (eBay about £19 inc carriage) as long as your element takes 10A or less.

I use several TET612 controllers, but mine are 240V and I use the SSR outputs.
It looks from the diagram that J1 is a relay with no and nc contacts. These should drive your two solenoid valves with no additional parts.

I would suggest using a PT100 probe. This will give greater inherent precision and the TET612 will read to 0.1C with it. When using a thermocouple, the TET612 will read to the nearest 1C and you have to be much more careful with calibration.

The downside will be the somewhat "lumpy" control. Being bypassed or going through the HLT coil will be very coarse and might benefit from a high flow rate to ensure that the returning liquor/sweet wort does not gain very much heat - this would result in de-naturing the enzymes and stopping conversion.
I personally use RIMS and raise the mash to 77C for mash out - this would be feasible for you if your HLT is at 82C or so.

You might have problems with the hydrodynamics - bang on/off of the two valves might shake things about a bit and possibly upset the pump.
I can see other possible problems with time constants but it would be a question of suck it and see - Good Luck!

You sound like really clever guy, is this your line of work? :thumb:

BB

 

[Aleman]

Can I just point out that several brewers I know who have gone down the HERMS route with the HEX in the HLT have abandoned the project. The issue is that the thermal mass of the liquor in the HLT the element cannot heat the liquor fast enough to compensate for the required temperature changes to the wort.

This is why people like Vossy, Seveneer and myself have gone for much smaller vessels as our HEX because the smaller amount of liquor responds much more quickly to the required heat demands.

 

[drut]

BB - electronics and associated control systems were in my line of work.

Aleman and others - I think you are missing the point. You do not need to change the temperature of the HLT to control the system. Please note my earlier point that the actual temperature of the HLT is relatively unimportant.

By varying the flow through the copper coil within the HLT you can add heat to the mash tun as required.
I am not convinced that a bang on/off system will be sufficiently controllable, but consider what would happen if you had a continually pumped system which either bypasses the HLT coil or goes through it.
An analogue analogy would be the output from the mash tun, via the pump, split into two and separately flow controllable. One circuit returns directly to the mash tun, one circuit goes through the HLT coil.
With the direct circuit open, and the HLT circuit closed - no heat added. With the direct circuit closed and the HLT circuit open - maximum heating - liquor/wort returns to the tun at 82C (but see my note earlier about flow rate and de-naturing).
With both circuits open, half the flow is heated, half not - very simplistically liquor/wort returns at say (82-66)/2C= 74C.

Therefore a PID with two, inverse analogue outputs, each controlling a variable valve would give continual control and would definitely work. Whether it would work well enough with bang on/off?? - probably if the flow rate and size (length) of the HLT coil were correctly scaled.

 

[drut]

I think it has to be said that Pjbiker has come up with a novel (to home brewing at least) and potentially very useful idea which some of us might use in the future.
Given that the HLT will almost definitely be on, full enough and at say 78-82C ready for sparge, there is no need for a separate HERMS assembly. I say again, the actually temperature and control thereof is unimportant.

Continuously variable valves are common in process control but are expensive. Some central heating zone valves are continuously variable and these might be used with a PID which has an analogue output. However, in the interests of cost, we can consider the following:

As described earlier the outlet from the mash tun is passed by the pump with its output split into two. These two routes should have manual variable (ball, gate) valves for setup purposes.
I do not think that there is a need for a solenoid valve on the direct return circuit as this will just make flow "jerky". The pump will operate continuously and with sufficiently large diameter pipe work, the total flow rate will be essentially constant.
The flow rate through the HLT coil should be manually adjusted such that the outflow temperature is no more than say 69C. This circuit is on/off controlled by a single solenoid valve such that heated liquor/wort is mixed with the direct circuit.
The overall effect will be almost identical to using a cartridge heater for RIMS but without the very high surface temperature which can cause burning if the flow becomes blocked. i.e. the heating will be added in pulses.

If my systems were not working extremely well, and I had the spare time, I would definitely try this.

Question - will a cheapo solenoid valve stand up to the frequent operation needed?
Answers on a twenty pound note to......

 

[pjbiker]

I think it has to be said that Pjbiker has come up with a novel (to home brewing at least) and potentially very useful idea which some of us might use in the future.
pulses.

To be fair, I stumbled on the solenoid valve idea on some american homebrewers own website I think, who had a similar but larger (20 gallon i think) set-up. My reasoning for keeping a constant pump flow is that if the pump stops, the temperature sensor in the mash tun inlet would be giving a reading for only that point on the circuit whilst the rest of it cools, and also should make the wort nicely filtered and temperature distributed throughout the grain.

Your extensive experience in the area is appreciated, thanks. My non Herms system does work pretty well at the moment and there's no real need to change, but it's interesting isn't it. I like to 'tinker'.

 

[unclepumble]

Drut,

The whole point of a herms is to be able top vary the mash temp, as well as keep the mash temp stable.

You can hold a mash very accurately at a constant temp for 90 mins very easily just by getting the tun insulated properly, without herms.

Herms is useful for protein rests and raising the mash temp for a a mash out as well as step mashing, therefore you need to be able to raise the temp of the heat source (a water bath with a heat exchanger in it) very quickly therefore small is more controllable, than large.

If you have a poorly insulated tun and want to keep the temp a constant running a coil through a large HLT will help, but if your PID is poorly set up and the Hysterisis is set too high about the temp set point it will take a long time to react to change and effect the mash temp as a result.

UP

 

[Springer]

Thank goodness for that UP, I was beginning to think I had been in a dream, living a false life, for the last year or more................ :? :? :?
I can now sleep well
:lol:

 

[drut]

Unclepumple - you are still misunderstanding the principle!!

RIMS & HERMS, both of which I do, on different setups, require the addition of heat in the return loop.

e.g. For a step mash - which I sometimes do, you start with the liquor/wort at say 50C.
You pass some of the return flow through the HLT at say 82C, therefore the mash gets hotter.
The amount of heat added is dependent on the percentage of the flow which goes through the HLT and the percentage which bypasses it. The heat addition is not dependant on the exact temperature, or controllability of the HLT - that is handled by the PID.

For mash out, the majority of the flow will be via the HLT coil, so the return temperature can be up to 82C.
This will raise the temperature of the mash to the required 77C provided that:
The HLT temperature remains above 77C and is not cooled significantly by the return coil.
The coil is sufficiently long, or the flow low enough to ensure that the temperature of the return is 77c or greater.

I have slightly simplified the delta T and the required use of the previously described manual valves. The "bypass" valve would probably be completely closed for mash out.

Please take the time to draw a diagram and consider how this works.
In HERMS you control the actual heat of the HERMS vessel, here you control the dwell time within the heated vessel and the ratio of the heated and unheated returns.

Springer - think it out for yourself! - I guarantee that it can be made to work - and I do know what I am talking about - it was my job. This type of system is commonly used in process control. You can't sleep yet!

I am now in the process of thinking through proportional control of the solenoid valve using the SSR output from the PID (with a buffer stage to get 8V at 40 odd mA up to 12V at 250 mA for the solenoid).
The mark/space is controlled by the PID and you can set the repetition rate on many PIDs.

 

[pjbiker]

I have slightly simplified the delta T and the required use of the previously described manual valves. The "bypass" valve would probably be completely closed for mash out.

In HERMS you control the actual heat of the HERMS vessel, here you control the dwell time within the heated vessel and the ratio of the heated and unheated returns.

I am now in the process of thinking through proportional control of the solenoid valve using the SSR output from the PID (with a buffer stage to get 8V at 40 odd mA up to 12V at 250 mA for the solenoid).
The mark/space is controlled by the PID and you can set the repetition rate on many PIDs.

Ratio of heated and unheated returns.
I can see how variable valves would work better. Didn't understand the 'mark/space, repetion rate' bit though.

You mentioned a simple thermostat controller for the HLT rather than PID, could you post an example? Is that because the control doesn't need to be so accurate?

Cheers

 

[drut]

The mark/space is the same as the on and off times. 100% mark = continuously on, 100% space = off.
50% mark/space gives equal on and off times. 33% is on for a third of the time.

The repetition rate is the frequency at which this happens, called "control period" on the PID instructions.
The default "control period" for N2006P and TET612 (two most common eBay PID controllers) is two seconds.
So when used with a SSR and with a mark space of 33%, the output will be on for 0.666 seconds, then off for 1.333 seconds, then on again etc. For lower heating, the "mark" will simply be less.
I say when used with an SSR because I am not sure that a solenoid valve will respond suitably if turned on and off this often.
However, if the "control period" is increased somewhat it may well work properly, hence my last two sentences above.

Temperature controller:
http://cgi.ebay.co.uk/Mini-Digital-Temp ... 2eb299a1f9

You might want to use a stainless steel through wall probe with this - it will need be a NTC type.
It is not that the control is less accurate, it just does not need to be as responsive. It is probably best if you do a search on PID Controller. This is a bit mathematical, but might help:
http://en.wikipedia.org/wiki/PID_controller

 

[unclepumble]

Drut, I don't need to draw any diagrams to think through your explanation, & I am not missing the point, (I currently work at the biggest Brewery in the UK, as an engineer in process brewing, so have deal with plenty of temp control equipment cooling and heating, daily, and I am pretty sh*t hot on large complex process machines having past experience on agglomeration equipment, powder blending lines, heat sealing equipment etc etc etc etc as well).

Your method will work best if you are controlling a normal cooling or heating circuit, such as IMS Glycol, brine or even steam.

However the wort in a mash contains enzymes, therefore if the liquid comes into contact with too high a heat you are going to denature those enzymes with the added heat.

Commercials don't use Rims or Herms they steam jacket the mash conversion vessel, and add steam to a large area to maintain or raise the mash temp, they also continuously agitate to give a homogeneous mash temperature.

Therefore my point is that if your using Herms you want to control the heat giving medium more accurately, rather than using a control valve or bypass loop, to ensure a consistent temperature in the mash and thus ensuring a more repeatable brew, that's why most people that go down the HERMS route, measure the temperature of the wort and hence control the PID using a PT100 on the outlet of the herms unit, as opposed to control from a PT100 in the mash.

Hope I have been clear enough or have I missed something :wha: ?

UP

 

[pjbiker]

Therefore my point is that if your using Herms you want to control the heat giving medium more accurately, rather than using a control valve or bypass loop, to ensure a consistent temperature in the mash and thus ensuring a more repeatable brew, that's why most people that go down the HERMS route, measure the temperature of the wort and hence control the PID using a PT100 on the outlet of the herms unit, as opposed to control from a PT100 in the mash.UP

I think I understand the point you're all driving at, and maybe getting 'crossed wires'. Please don't get offended folks. The intention was to use a PT100 in the outlet of the herms coil, before it enters the mash tun.

Can I just point out that several brewers I know who have gone down the HERMS route with the HEX in the HLT have abandoned the project. The issue is that the thermal mass of the liquor in the HLT the element cannot heat the liquor fast enough to compensate for the required temperature changes to the wort.

This is why people like Vossy, Seveneer and myself have gone for much smaller vessels as our HEX because the smaller amount of liquor responds much more quickly to the required heat demands.

I think Aleman has a point here, the limitation is the efficiency of heat transfer between the heating source and the circulating wort. My plan was to use a length of 10mm copper in the HLT at ?>77deg C. Perhaps as he says, this is not efficient enough a transfer to increase mash temperature in the timeframe required for a 'mashout', regardless of the flow-rate through it.
My plan was more for a maintenance of homegenous mash temperature, but would obviously like the flexibilty to try stuff out. Its till a 'pipe dream' pardon the pun.
Peace out.
PJ