Build yourself a cheap digital boiler power regulator

The Homebrew Forum

Help Support The Homebrew Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Good to see this thread back to live. I used to use one of these (see the picture below) until it failed on my last brew day with a smell of burnt electronic component from the box.
1620645997426.png


Now I'm wondering if it's a good idea to replace it with this plus an active in-box cooling fan.
Any ideas are highly appreciated.
 
The thing to watch out for is the power dissipation.

These devices control the power by using an electronic switch called an SCR (aka thyristor) or sometimes a triac.

It's an inescapable characteristic of such devices that they drop about a Volt across them when they are turned on. This means that they need to dissipate 1 Watt of heat for every Amp of load current passing through them.
To put some context on that, 1 Watt of heat is enough to raise the temperature of 1g of aluminium by about 1ºC per second.

These units typically claim to be able to control 4kW - i.e. a load current of 18A. That means it has to dissipate up to 18W of heat (at full throttle).

If you look at the spec for heatsinks (e.g. on RS) they are rated in ºC per Watt, which tells you how much hotter than room temperature they'll get for a certain amount of heat power.
Typically for the sort of size heatsink shown in the product photo (100x60mm) you're looking at around 3.5º per Watt, assuming good ventilation. So at 18W you're looking at 60ºC above ambient - say 80ºC.

Personally I'd say a cooling fan was absolutely essential from a safety perspective; and I'd be rather sceptical about the ability of a device like this to handle 80A for more than a few seconds.

In my humble opinion I would avoid using devices like this to control anything more than a couple of Amps. A telling question to ask the seller would be "what is the maximum continuous load current that the device can safely deliver using the supplied heatsink?"
 
Last edited:
Good to see this thread back to live. I used to use one of these (see the picture below) until it failed on my last brew day with a smell of burnt electronic component from the box.
View attachment 46860

Now I'm wondering if it's a good idea to replace it with this plus an active in-box cooling fan.
Any ideas are highly appreciated.
Personally I'd steer clear (see above)
 
A properly ducted cooling fan of the appropriate size & airflow makes an absolutely tremendous difference to the ability of a heat sink to remove heat from the device being cooled. Loads of good technical info online.
However the problem with cheap & nasty electronics is that the device itself sometines isn't thermally bonded very well to the heatsink. But overall a good supply of cool air should improve things greatly and increase long term reliability too.
 
@The-Engineer-That-Brews , thanks for the explanation. It's really helpful. My background is in mechanical, so I normally steer clear of any electronics but in this case I'm not sure what would be a safe alternative.
I have a 3kW tea urn which boils way too vigorously. An SSR with the appropriate heat sink will probably do the job but it needs a rather complex (for my skill set) control circuit. I guess a could use an SSR with resistance control but I couldn't source one. What else can I use?
@Bill_g, appreciate your point, a good airflow is a must.
 
I notice that fully assembled but unboxed 8kW power controllers are available either on eBay, Amazon or elsewhere. (searcH for '220V AC 8000W controller'. These devices are pretty simple in electronics terms. Rotary or digital controls are available. If I was doing this I'd just buy one & do some thermal & voltage tests. (Possibly a little tricky if inexperienced plus high power mains is a bit unforgiving.) Generally speaking the Far East can manufacture fully assembled units considerably cheaper than we can buy just the components in the UK.
My expectation would be that a low cost device rated at 8kW ought to manage 3kW continuous. However tests are the only thing that would tell whether the heatsinking is really adequate. As mentioned previously the addition of a small fan would make a very large difference to heat sink efficiency, but it may well be unnecessary if the heatsink is mounted in optimum orientation inside a well ventilated case to encourage good convection.
 
@The-Engineer-That-Brews , thanks for the explanation. It's really helpful.

I guess a could use an SSR with resistance control but I couldn't source one. What else can I use?

You’re welcome - sorry if I sounded a bit stuffy, I was trying to do too many things at the same time (!)

SSRs are really just on/off devices and can’t be ‘part on’. The way that they are used for power control is essentially just to switch them on and off with a given duty cycle so as to deliver the desired average level of power to the load.

From what you wrote, I understand you‘re looking to control the amount of power to your boiler to a certain set level (say 40%), rather than to hold it at particular temperature (for which you’d use something like an Inkbird or other cheap PID controller). This kind of power control is something I do myself with my Berco and it helps reduce boil-off (and sometimes boil-over...) quite a lot.

Unfortunately doing this with an SSR requires a bit of additional circuitry and that’s a route best avoided (as you wisely say) unless you know what you’re doing.

As an alternative, I’d suggest wiring two or three 100W light bulbs in series with the element - with a switch to bypass them when you want full power. If that doesn‘t reduce the power level enough, try using a 1kW electric heater element (suitably mounted of course).

Hope that might give you a possible route ahead :-)
 
As an alternative, I’d suggest wiring two or three 100W light bulbs in series with the element - with a switch to bypass them when you want full power. If that doesn‘t reduce the power level enough, try using a 1kW electric heater element (suitably mounted of course).

Hope that might give you a possible route ahead :-)
[/QUOTE]
As a retired electronics engineer of 40 years experience I'm utterly speechless. Please no one do this. Appalling advice.
 
@The-Engineer-That-Brews , I hope your light bulb suggestion was a joke and I really appreciate your sense of humour :)
Even though my engineering degree is not in electrical/electronic, I have enough common sense not to start wiring lightbulbs in series with a heating element. I don't like the idea of another heating element in series either. Even if electrically okay, these options raise some other concerns.
So I can see my option as follows:
1) Buy the Chinese 1000W power controller (to be used for 3kW load) with a cooling fan. It will have enough margin, I guess;
2) Read a bit more on how to wire a Raspberry Pi as a PWM controller for the SSR (I know how the SSR works and I know how to code the PWM, just need to safely wire the RPi GPIO pins to an optoisolator and the SSR);
3) I found that my 3kW heater is actually 2x1.5kW heating elements in parallel, so I can wire a switch to disconnect one of them.

Number 3 looks like the obvious non-electronic choice but is the least desirable one as it doesn't provide a smooth control.
 
Heh - well as it happens, my degrees are in electronics and electrical engineering; and when I was doing my power engineering course, using incandescent light bulbs as mains current-limiting elements was quite a well known technique - but perhaps you've not encountered it.
 
I can't see what's wrong with the light bulbs and heater apart from not being able to buy them anymore and the the cost of the electricity you are then wasting, but from safety aspect what is the issue ?
Yep - a lightbulb in series used to be a standard technique in the lab for testing dodgy old kit that might have a short-circuit in the power supply.

You're right about the power wastage of course, and it's certainly not the way I would choose to do it (I use SSRs chopped on and off under MCU control); but as the OP was asking for alternative ideas that didn't involve them building a circuit, I though it was worth mentioning.

I'm assuming the power controller he proposed using is a phase-angle controlled SCR - the electrical noise from that with a 3kW load must be something else...
 
Found an interesting youtube video stripdown of what looks very much like the 4kw power controller's insides -

Looks like a 10A fuse inside so only 2.3 kw in reality though I guess you could up it to 15A if you were confident you'd got enough cooling air.
 
2) Read a bit more on how to wire a Raspberry Pi as a PWM controller for the SSR (I know how the SSR works and I know how to code the PWM, just need to safely wire the RPi GPIO pins to an optoisolator and the SSR);

Oh if you're happy using a Pi, then controlling an SSR from it is a doddle - the SSR modules are opto isolated internally, so it's really no more complicated than controlling an LED (but obviously do make sure that you keep everything away from the mains terminals at the other side, and USE A FUSE).

You DO want to make sure that you buy the right sort of SSR however, and that you get it from a reputable supplier. I'm afraid that every single one I've bought on Amazon or eBay are fakes, and these days I will only buy them from places like RS, Farnell or DigiKey - they do cost a few quid more but it's worth it.

You want to make sure you buy an SSR with enough current handling capacity, that is (a) optically isolated; (b) zero-cross and (c) has an input voltage range that is compatible with your PI.

A good example is these ones from RS (about £16 with the vat). I've got quite a few of these and they work absolutely fine controlling 3kW heating elements. Honestly they are really easy to use - the corresponding RS heatsinks are very good too, and will allow you to do a 2hr boil on a Burco without any additional fan cooling.

Good luck and play safe - if you need any help or advice hooking it up do give me a shout.

Screenshot 2021-05-11 at 11.26.45.png
 
Last edited:
As a retired electronics engineer of 40 years experience I'm utterly speechless. Please no one do this. Appalling advice.
[/QUOTE]
And your reason is...?
[/QUOTE]
Oh come on, you only have to do a few simple sums to realise that it won't work. Anyhow resistive power limiting went out of fashion when Noah was a boy.

The Youtube analysis of the '4kW' power controller is excellent & he did well to find the explanation of the crafty use of the bridge diode assembly. The inclusion of an internal 10A fuse indicates that the device is rated at AC peak 4kW not RMS. It will probably support 2kW continuous but not much more.
If as mentioned above your heater element has 2 x 1.5kW sections then you could use two of these devices as they are very cheap. It's unfortunately more costly to buy the components than to buy a whole assembly from China.
I note that 8kW versions are available via usual sources (eBay, Amazon etc) which might feasibly support a 3kW load comfortably. But as always the acid test would be an experiment on such cheap (& sometimes) nasty electronics.

On the topic of electrical noise which was mentioned. This will mostly be in the low frequency bands and highly unlikely to interfere with any modern devices. But I expect that the 'Engineer Who Brews' listens to Mantovani on his long wave valve AM radio where there could certainly be a noise issue.

I don't know why some above seem to favour the use of logic circuitry for 50Hz phase angle control. It would be a very complex solution to a very simple analogue circuit design as shown very well in the Youtube video.

Ffinally, the main advice I'd give to any electrical novice is make sure that your earth wiring is good & that all exposed metal parts near to your circuitry are earthed. Always use an earth leakage trip. Those precautions will limit the possibility of a nasty accident.
 
I think I need to unsubscribe from this forum, I can't take any more!
 
Back
Top