I built my HERMS from scratch, including the control unit (I'm an electrical engineer). A schematic of my system is below:
I start by heating my strike water in the lauter tun. The lauter tun is a 64l aluminum pot and it is fitted with a 1500W electric heating element. The lauter tun also contains a temperature sensor. When my target temperature is reached, I pump the water into the mash tun, via the heat exchanger. The mash tun contains the dry grist before the water is pumped in. I find that I don't get any clumping (starch balls) at all with this technique. I always choose an initial strike water temperature which will result in a mash temperature which is below my target because 1) I can't be bothered to try and hit it dead on, and 2) the system gets it up to the proper temperature quickly enough for my tastes. I actually strive for a rather low strike temperature when I brew wheat beers, as the time the mash spends at the lower temperature is a rather good protein rest (for my tastes anyway).
The heat exchanger also contains a 1500W electric heating element. The HE contains a (roughly) 8m long coil of 9.5mm ID copper tubing - the ID of all the other hoses in the system is 12.7mm. The reason for this mismatch is that 1) my pump's fittings are 12.7mm and 2) the coil is actually my recycled immersion chiller I had built many years before. I didn't want to buy a large quantity of 12.7mm ID copper tubing when I built the system when I had an acceptable alternative on hand ($$$). The HE holds 6l of water.
The reason I employ 1500W elements and not higher power elements is that I'm powering the entire system with ordinary (for north america) 120V outlets. I could have designed the system for higher power 240V elements but then I would have had to tear apart my basement to put in a dedicated 240V circuit for it. I couldn't be bothered with doing that, especially given how much I hate doing drywall repairs.
The mash tun is fitted with a false bottom and also contains a temperature sensor. It is also a 64l aluminum pot. All vessels and hoses have been insulated.
Once the initial strike water has been pumped into the mash tun, I set the valves to recirculate and start recirculating. The system then brings the mash temperature up to the target and holds it there. I hold this temperature for 60 minutes, then the system ramps it to 75C for mash out. Once mash out has been reached, the mash tun is drained to my kettle. While this is occurring, my sparge water is heating to 75C in the lauter tun. When the mash tun is drained, the sparge water is pumped into the mash tun. Once the water has been transferred, the system is set to recirculate again. I recirculate until the mash temperature comes back up to 75C, then I drain into the kettle again. This is a modified batch sparge technique - I don't fly sparge. I brew ~40l batches and an typical 1.050ish OG brew requires about 63l of water, which I split about evenly between the initial infusion and the sparge. I normally get 85-90% efficiency for a 1.050-1.060 OG brew. Efficiency drops when the OG rises above about 1.060 or so. I get about 70-75% for a 1.090ish OG.
The extra valves in the lines dowstream of the lauter and mash tuns were installed so that each vessel could be removed without having to shut down or drain the system.
I don't measure the temperature at the outflow of the heat exchanger because 1) given my system it's a pain in the ass to do and 2) I'm more concerned with my bulk mash temperature. I know some of you will point out that my outflow will be hotter than my main mash temperature but given that I measure the temperature of my HE, the outflow can't be higher than that. During normal operation, the max delta T is approx 2 - 2.5C. I can live with that. The temperature sensors are each mounted in a thermowell (hollow copper tube). The sensors are LM35 ICs, and can't come into contact with liquid. The thermowells aren't permanently fixed into the vessels; aluminum & copper don't "like" each other and once again, I just couldn't be bothered installing them permanently.
The lauter tun's lid is fitted with a motorised stirrer. Nothing fancy, just a scavenged electrical motor bodged to a piece of coat hanger. I use an ordinary dimmer switch to control its speed. When I first built the system I included a stirrer in the HE, but given the heat and time it was running, the motor seized. To be honest, the system is actually more stable since the HE's motor stopped working. What I mean by this is that the controller hits and maintains the set temperature with less of an over/undershoot. In fact, it's practically nonexistent now.
I do not have a mash stirrer, nor do I need one. The wort recirculation achieves the same thing.
I wrote an article about my system for a magazine a few years ago. I think that posting the article is probably contrary to the rules but we'll just keep that to ourselves, shall we?
Here it is:
View attachment 2606016.pdf
For those of you inclined to create your own PID controller, the excel worksheet I created to help me tune the system may help you:
View attachment pid_workbook.xls I also recommend that you read
PID Without A PhD by Tim Westcott.
I think I'm near my post limit regarding attachments, so I'll save the pictures for the next post.