Zizzle's Brewbot

[ArnieW]

Something else useful I just stumbled upon: http://www.acroname.com/robotics/parts/R48-IR12.html

Hi Zizzle,

I spent a lot of time with one of these - they do not reflect well off wort or liquid; and I even tried a float system and got terrible reflections that made it very difficult to use. Ultrasonic is much more reliable in my experience.

cheers, Arnie

 

[Zizzle]

Good to know. What doesn't work is very valuable info.

Have you tried your flow switch yet? I got one too the other day, but only got home from a weeks holiday last night. Itching to test it out tonight, but SWMBO said it was time for a food shop. And I've got a beer to bottle for the case swap. Bah.

 

[WildaYeast]

Freakin far out Zizzle. Read a bit a while back and just caught up. Will have to come by to see it. Just did my first AG over the weekend (finally, long time since Brads). First was a stout, but I've got grain for a skunk fart as well -- same as we did that day. Will get it going soon hopefully and arrange to come by wiht some when ready, if not sooner.

Brian

 

[Zizzle]

My flow switch has been plumbed in for a week now, software tweaked ready for a test.

Not a great photo, but the flow switch is the short piece of copper with a black blob on the side.

So I finally ran it up for a test. Seemed to work well enough. Ideally the flow switch would only activate at higher flow rates. I think it's calibrated for 0.5 litre/min which is just a trickle. It doesn't seem to restrict the flow much at all.

But with some more software tweaks I managed to get brewbot through a dummy test brew without a hitch. For the first time.

I have a simple and obvious algorithm implemented: run the pump for a few seconds, and if there is no flow, shut the pump down before trying again in a few seconds.

Usually after valve setup gravity will put some fluid in the pump, but some air pockets will prevent the pump from working. So running the pump for a few seconds then stopping is usually enough to dislodge any air pockets in the system, and the second try will see the pump primed and flow start.

There is also some logic around bailing out if it tries to start the flow too many times, and also not bothering to try to restart the flow after a decent period of transfer time.

Of course it's yet to be see if it all goes without a hitch when grain and wort is involved.

 

[Zizzle]

Some slight reconfigurations today.

The main issue is to make it easier to drain water out of the system after cleaning.

So I've moved the valves. There is a slight slope to the left so I can easily pull the silicon hose to the HLT and have the fluid drain out:

Hopefully soon I will get the point where I'm convinced I won't need to do anymore welding and can paint the thing and make it look pretty.

Yet to do: Make the hop dropper a little more robust. Mount some fans to push steam away from the hop dropper.
Then on to integrating a HERMS setup.

 

[InCider]

Great rig Zizzle :super:

Bring it to Sqyre's at Xmas and we can have an AHB Summernats!

InCider.

 

[Zizzle]

Sure, what sort of rig will you be sporting at the summernats?

A fully sic Turbo-rotary-V8-diesel-on-gas-with-four-on-the-floor-three-on-the-tree-and-a-sixpack-in-the-back?

 

[bonj]

My 3 DS18S20+ samples arrived today... They took their time, but at least they arrived!

 

[Andyd]

Mine came today, and last week

I resubmitted my order a few weeks back when I figured my first order had gone through the gaps

Andy

 

[bonj]

Mine came today, and last week

I resubmitted my order a few weeks back when I figured my first order had gone through the gaps

Andy

I thought mine had gone through the gaps... but they turned up. Will be fun playing around with these.

 

[Zizzle]

Long time no post, but there has been quite some progress.

To solve the hop dropper problems with steam I wanted to enclose it in a case a have a small fan pressurise it. So I got a cheap plastic box and built a new dropper setup out of some scrap MDF and one of my wiper motors.

A while back it pissed down and poor old brewbot was wet for week out on the deck. It started to rust.
So I was force to paint him.

While stripped down, the brain got an upgrade to a set of 4 triacs with opto-isolators I had laying around. To fit them in with PCB & heatsink, I had to move the CPU board & transformer.

I have not moved the heating element control over to triacs yet. But the pump is on a triac, and I have another set aside for running a HERMS element. For those not in the know, a triac the main component of a Sold State Relay. It will allow my to drive the 240v get at a variable duty cycle, like a dimmer.

I also spent a lot of time tracking down some issues with driving the wiper motors.

The problem was that the CPU would sometimes reset when changing the valves. Or sometimes it would latch up hard, with only a power reset bringing things back. So I wrote a little test harness to cycle the valves. It would only survive a few iterations, so it was quick to test attempted fixes.

I was pretty sure it was a hardware problem. First stop was some more filter caps on the power supply. No change. Then a zener diode and filter cap on the motor-stall current sense circuit. No change. Then I tried disconnecting the current sense circuit. No change. Then I tried shorting out all the motors to prove it wasn't a power supply glitch. The test would run indefinitely proving that the power supply had no trouble delivering the current needed: the problem was clearly due to the inductive nature of the load.

I couldn't see why the CPU was having trouble. I tried running it from a separate power supply, but still the resets were happening. Finally I found a bipolar electrolytic in the junk box and put that across one of the motors. Problem solved. On the CRO I could now see the spike. It was still around 80v, so it must have been much higher undamped. Not being a digital storage scope it was hard to see, and undamped impossible.

It seems that the inductive flyback from a stalled wiper motor is enough to somehow escape the relay, find it's way out through the driver transistors back on to the CPU board. That means pretty high voltages... maybe over 1kV. I'm just glad I didn't fry the CPU or any of it's pins.

All together now:

The software and UI got some bugs fixed, and other upgrades. Hopefully a will do a brew tomorrow night.

 

[Zizzle]

Tonight I stupidly walked away while testing out some new code that runs some of the steps in parallel (e.g. heating HLT to set-point while recirculating). The code was buggy and applied power to the HLT when it didn't have water in it. I smelt something was not happy, ripped the lid off the HLT to see both the elements glowing yellow hot. Hit the reset button. Remarkably they both survived, although now have a nice shade of blue. You can't beat the $9 kettle element:

I don't know how this may have effected their life span, but it sure did get all the gunk off them.

 

[grinder]

Hey Zizzle.
Are you an Electrical engineer of something like that? How do you know how to do all this? Where can I learn how to do it?
It's genius!!

Grinder

 

[poppa joe]

WOW...
But i think the yellow wire is the problem...
Cheers
PJ

 

[Zizzle]

Yeah, I have a Systems Engineering degree following a mechanical engineering stream, but I work in software engineering (mostly self taught), and all the electronics engineering was self taught (during high school).

But mostly you've just got to be motivated to have a go, learn, fiddle around, and experiment.

Pick up some books on electronics. It's pretty easy to pick up and you can even avoid the complicated maths if that is not your thing. The internet is a great resource and I wish I had it when I was learning. Most of the time circuits for what you want to do have already been designed. It's just a matter of building them and integrating them with what you already have.

 

[Steve]

You are a Class A Nerd! In a nice way of course. Thats amazing.....like I said earlier in the thread. No idea half of what you're talking about but **** its good to see all your work/hobbies/passions coming together to make beer :beerbang:

Cheers
Steve

 

[devo]

you know if you had painted it red it would go much faster!

 

[Tony]

I want one of these but i will tick the brewery option box instead of the BBQ option box

you could thak that to summernats

cheers

 

[Sammus]

I want one of these but i will tick the brewery option box instead of the BBQ option box

you could thak that to summernats

cheers

:lol: I think that spit might spin a little too fast. I guess it could be geared right down to turn an elephant or something :blink:

 

[Zizzle]

Short video of me testing the hop dropper tonight