I've had mixed results recently working on HERMAN. Last week I began the process of re-plumbing the solenoid valves to tweak machine performance. This included having two water in valves (one dedicated to the mash tun and one dedicated to the HLT) after I found that reverse pressure would open them. The other change was brought about by installing the immersion chiller with agitator in the kettle. This meant I no longer needed to circulate kettle wort through a chiller until it reached pitching temperature. A side effect of this was that there was only one wort path out of the kettle (into the fermenter) compared to two previously.

I had to pull apart the valve that was part of the old kettle wort chilling loop and discovered a good deal of hop debris in it. I think it was a good move to eliminate this path. Now the chilled and settled kettle wort simply gets dropped via gravity into a fermenter. The valve and fermenter fill path will be sanitised at the start of brewing with Starsan or equivalent.

Through various plumbing changes I had a significant amount of trapped liquor. Revising the solenoid arrangement gave me the opportunity to add a low point valve (manually operated ball valve) for drainage. The changes also meant that the solenoids needed re-wiring.

All that seemed to progress quite well until it was time to connect everything up and test. The manual switches did not engage the relays and after about 15 seconds puzzling I noticed that a ribbon cable connecting the BrewTroller controls to the relay card was melting! I have no idea yet what was happening but this is one of the drawbacks of creating a complex vero-strip interface. I noticed one cable was not plugged in, but not sure why that would make a difference. At least the BrewTroller seems to be fully functional so that is a blessing. I'm eager to get a version 3 board with 16 outputs (assuming that will be the final arrangement) so that I can settle on some final interface designs and etch decent relay boards.

So at the moment I have a pretty looking set of solenoids that are not working.

Beyond that there were numerous small tweaks. This included improved plumbing between solenoids and tuns and changing the depth of the kettle temperature probe so that it is higher in the tun than the heating element. The idea is that when hot wort from the mash hits the probe, the BrewTroller can automatically apply kettle heating because the element will be covered. The BrewTroller control box was finally mounted properly with suitable placement of cables and space for the laptop to sit if so desired. I think pretty much every time I brew I will have the laptop running with BTremote just for the convenience of it.

I've managed to stockpile enough beer to last me through a longer phase of BT improvements so I'm in no hurry but rather wanting to apply good solutions rather than temporary ones. Assuming the relay interface board is actually ok, I think I will brew in two or three weeks. I'm aiming for a nice tasty American Pale Ale.

Beyond physical construction work, work has continued on the BTremote program. There have been a lot of tweaks communicating between BT and the PC and I'm nearly ready to release an update where recipes can be done reliably on the PC and downloaded to the BT. This means no more painful tweaking of the rotary encoder to add recipe names and set parameters. It is a great input option for a stand-alone device, but so much easier if you don't need to use the encoder for detailed input like setting up a recipe. My APA recipe was uploaded from the PC to the BrewTroller without even a tweak of the encoder, so a big step forward there.

A few months ago I uttered something about actually finishing a brew rig before I updated to the next model. OK, so I could not help myself once the BrewTroller arrived. I never did finish working on HERMAN 6. I didn't finish coding for it. I never did design and etch a motherboard for it. And I never did get to a final plumbing layout.

But I did learn a lot along the way.

And all that experimenting has given birth to a new machine, and this time one that is radically better than all previous incarnations.

For a start, I'm not trying to do everything myself this time round. The BrewTroller is a good example of what I mean. Matt and Jeremiah have put a lot of work into a sanguino based brewing controller that is open source. The base model is full of features I'd barely even dreamed of, and I still get to code my own bits into it if I want to.

Also, having a like-minded brewing community means that ideas and resources are shared more easily. I have taken the plunge and bought cheap solenoid valves from China that so far look to be the goods.

And having solenoid valves available means extra features are easy to implement. I've added a mash idle configuration to the plumbing layout as an example of this. Until now when the mash reached temperature (or went too high), all HERMAN 6 could do was stop recirculating. Now the mash idle valve means that the heat exchanger can be bypassed, meaning that mash liquor is always recirculated, and that overshoots will drop to normal more quickly.

I didn't mean to kill HERMAN 6

Once I got the BrewTroller I did some experimenting to get to know its capabilities. Then to test it I needed to add pressure sensors, which I did. Then it made sense to wire the temperature sensors into the BrewTroller. And then I needed to test outputs so I built a relay board. And then I thought that it was time to buy zero-crossing solid state relays to power the hot liquor tun and kettle. And by then ... HERMAN 7 was so close to being a reality that it made no sense to hold onto nostalgia or legacy bits and pieces.

And behold ...

Before long the solenoids went in. And then it was a simple matter to wire in the relay board. And then the solid state relay board simply plugged in. And by then ...

... there was no more HERMAN 6 at all.

A quality upgrade

I've had the luxury this time round to do things right. This has meant buying quality components rather than cobble together bits and pieces that kind of work. I haven't gone to the degree of building everything in stainless steel (that remains beyond my skill level), but all the fittings are quality fittings, including stainless camlock fittings throughout the rig.

So close ...

There is still a small amount of wiring to do. I am short a few plumbing bits that I will need to get via mail order. I am short a few switches and LEDs to do a complete dry run. But by the weekend I will be ready for a wet run, and then the weekend after I expect HERMAN 7 will be making beer.

Even though it has been ages since my last blog entry I've been busy building and inventing. One of the things that has taken my attention is putting together a BrewTroller control panel.

There is more information at the BT forum here.

The fridge tech confirmed that things were not good for the failed glycol ice bank. At best estimate it was going to cost $750 to repair, and possibly double that. I had anticipated it might not be cost effective to repair but I was not expecting it to be that bad ...

So I had been cooking up an alternate plan. I'd been eyeing off small freezers selling on ebay. A factory second or a used one ought be suitable as all I am after is a cooling plant. I intend 'skinning' the cooling plant anyway, so cosmetics are of no consequence at this point.

While the glycol unit was an exciting addition to the home brewery, I realised that a small freezer was likely going to be more economical to run and definitely quieter. The other advantage is I could get a freezer that was the same height as the 'cellar', which means the bar top will be flat now rather than stepped.

There are quite a few bits from the old unit that are salvageable. Most importantly, the stainless cooling lines that are submerged in the glycol will find their way into the new system. They are effectively form a heat exchanger - beer flows from a keg at cellar temperature (say 11 C), and while passing through the stainless coil submerged in glycol at around -3 C, it is flash chilled to 4-6 C when it emerges from the faucet.

The next useful bit that stays is the flooded font itself. This is a three tap system which will ice up nicely if the glycol is cold enough - a good way to impress the neighbours.

And the other crucial bit that I'm salvaging is the controller which will regulate the glycol bath by temperature controlling the freezer power. Oh, and there is the stirrer as well which is used to keep the glycol and water mixed. Without it the glycol tends to settle at the bottom of the tank.

Meanwhile work has continued on the cellar/bar combination. The mini corrugated iron has been added and a wooden strip to one side now houses the two control units - one to set the cellar temperature via pumped glycol and radiator with fan, the other to set the glycol temperature itself.

The cellar we are making is simply a glorified 'Son of fermentation' chiller. It differs in a number of ways though. Firstly, it is heavier and larger because the insulated box is an old refrigerator. Secondly, the cooling is more automated. Instead of dropping in frozen bottles daily, the HERMAN cellar is using cold glycol from our glycol chiller to provide the cool air. To convert from liquid into cool air, we are using a PC cooling radiator.

The compressor shelf in our old freezer was taking up too much room. With the shelf in place there was only room for 4 kegs. By removing the shelf and patching up the holes, we could fit 8 kegs in - a worthwhile difference.

Cutting the old shelf out was much more difficult than I expected. It was certainly easier creating the new wall cavities and filling with insulated foam than it was destroying the old bits. The photo to the left shows what we started with once the shelf was removed.

We needed framing timber in place before we could skin the wall. The existing foam insulation was trenched neatly with a dremel. The job was quite messy and a mask would be good if you plan to repeat what we have done. The fine bits of foam are best vacuumed for clean up.

The framing is beginning to take shape. It was pretty much test and re-test and adjust until things seemed to fit. The PC radiator is sitting in close to final position. With glycol pumping through it and a fan forcing cold air into the cellar, we are confident it will cool the space nicely.

The framing means we can attach thin sheeting on the inside to skin the inner wall, and patch up the outer holes for the outside wall. The cavity is then filled with space expanding foam which gives both structural strength and has insulating qualities.

My mad foaming assistant is ready for the job. This stuff keeps expanding for a long time after it has entered the cavity. We had quite a lot find its way out of the smallest of holes resulting in funny foam sculptures. As long as it insulated the whole box it will be worth it.

I tested the radiator and it will definitely blow cold air if the water is cooler than ambient. I didn't record any actual figures, but anecdotally it will easily provide the cooling needed.

The font flooding pump and lines were given a test run after new seals were added to the font. All seems good there as well. With new glycol on order and the pretty skin for the cellar ready before next weekend, it seems the next major thing is regassing the glycol chiller.

Oh, and the cream ale and American wheat are both looking promising. We will brew a cross between a cream ale and pale ale this weekend - if you like it is a milder version of the pale - citrusy but with some subtlety.

The bar and fermenting areas have long been the most neglected parts of our home brewery. I guess I've been too fixated on building the HERMAN machine to put the time and effort at this end of the process. From time to time we have paid the penalty for it, with some beers fermented too warm or with spikes in temperature. The fridges we have to ferment and serve are nasty little bug factories as well, so it is time to sort these problems out.

Last blog I showed a mock up of how things will go together. Now for the serious work to put the bar together.

The photo above shows the framework that is needed for mechanical strength. This is the stuff that hold the overhangs in place, gives support where the flooded font is mounted, holds the drip tray in place, and allows clearance so the side door can open and close. The framing is built from 40x18mm pine. To make sure things were going to fit, I taped the bits together.

The font needs a 50mm hole. Two pieces of the pine provide the mounting support and the hole was drilled with a forstner bit.

I was going to dowel joint the pine framing together but realised that this was overkill. They are simply butt glued with Araldite, and this is sandwiched with 3mm particle board below and 7mm Jarrah flooring above.

With the Jarrah top in place things were looking good. Once the sandwich was glued, the edge trimming was cut to size and glued using packing tape to keep the strips in position. These were stained a Jarrah colour, and the whole thing finished in a high gloss floor finish.

While I was doing all this, Ani was out at a local market and came home with a sign for the new bar. And below we have a near finished top-side of the bar. There is work to do on putting a nice looking skin on the old freezer, and there is quite a bit of work to do on the cooling side of it as well.

On Monday I tackled the old freezer that I am going to use as my 'cellar' storage and fermenting area. The compressor shelf took up a lot of room, so the idea was to cut it out and then re-seal it as an insulated box. The cooling is to come from a separate glycol chiller and a PC cooling radiator.

The photo above shows the hack job in progress. I'd figured this would be a 2 hour job, but it actually took the best part of the day. They are not designed to rip apart. It was a lot more solid than I'd imagined.

With the angle grinder, a hack saw, power nibbler and various miscellaneous tools, the box began to take shape.

In the distance you can see the closed 'lid' of the freezer. This will be the access door for kegs and fermenters. The lined wall can be clearly seen in the photo minus the compressor shelf. I will be building a new straight wall and filling it with a foam expander to act as insulation. The radiator will be fitted into this fabricated wall.

It is not yet pretty, but a concept of the final product. The top of the box will be made with Jarrah flooring (a deep red Australian hardwood), and coated with a high gloss floor finish to create a durable bar. I will put a timber finish on the front of the box as well to pretty it up and get away from that whitegoods look. If the glycol chiller can be fixed, it will get a similar top that will also hold the English beer engine. I will put both on castors so that it makes the whole setup easier to move.

At the HERMAN end of the brewery, I've finally completed the LCD installation and went to connect my usual laptop to the picaxe via a USB-serial adaptor. It appears the adaptor has failed . So I've spent an extra hour or so tonight getting an old machine with a serial port loaded with the latest software so I can use it to code the LCD display.

I have both Monday and Tuesday off this week, but Monday will be spent down on the coast with friends. On Tuesday I'm planning a double brew to catch up. Currently we are a house with no beer . I'll be brewing a cream ale and an American wheat. Both ought be quick and easy to ferment out and light and easy to drink as our weather warms up. After that I intend to brew some more lagers once the 'cellar' is ready to go.

It has long been our goal to be able to control HERMAN remotely via the internet. There have been some unexpected twists and turns in the last week that have pushed us a little closer to realising the dream.

I've already mentioned how easy it was to set up a remote desktop on XP professional. This week I have been doing some VB Express 2008 coding, following Dr Acula's instructable on using FTP to connect two machines together.

I got there through an unexpected route though. The control panel had its finishing touches done on Monday. At the same time I wired in the backlight for the LCD, and the -12V line from the PC supply to provide a negative contrast voltage (around -1V). When all this was done, two things became apparent. Firstly, the old 40X control board from HERMAN5 was no longer working, and secondly, the LCD is blank and I still don't know why.

I considered spending the time to work out what was wrong with the 40X board. It has always been trouble, mainly because it was built double-sided with poor alignment, etched poorly and then drilled poorly. Board through vias have been hand soldered onto pads that only partially existed and did not match from one side to the other. I already had a 28X1 picaxe running on a breadboard, and considering it was a testing prototype for what will ultimately be the HERMAN 6 control, it seemed the perfect time to push it into service.

There are other good reasons to go this way. The X1s are far more powerful than the Xs. Using onewire protocols, the temperature probes can be read without implementing a 750ms delay. A five second delay with all the DS18B20s that I use means the old picaxe is incapable of anything else during those read times. So if we wanted a refresh on temperatures every 5 seconds, it would literally cripple the processor. The X1s can read a temperature while performing other tasks. There is also a timeout on serial input reads, which has been a problem to workaround on all other picaxes to date. Anyway, this is the path I've chosen, and so a good deal of time has been spent working on code to regain control over the machine ...

It is hard to control HERMAN without a display. For the moment I have chosen not to worry about sorting out the LCD, but rather work on a simple interface between the picaxe and my laptop, and use the laptop as the display. This was always part of the plan in any case, so like the 28X1 breadboard, it seemed like the right path to take.

I had been aware of Dr Acula's excellent instructables for some time, but had not made time to test the process for myself. It has been fun to work through the tutorials, and very exciting last night to get two PCs to remotely talk to each other via FTP. The matrix is coming together at last.

All I have to do now is finalise coding for the data sharing between picaxe and PC, and also the picaxe coding to switch SSR outputs. I imagine this will be completed by the end of next weekend, and I'm still waiting on the stainless fittings for the mash tun so I can't brew anyway.

It has been a satisfying couple of weeks. There is only so much creative energy that can go into planning and virtual testing. We have been building the circuit boards for the control panel and having a lot of fun making things.

The photo below shows our Dremel in the #220 drill press. Having tried to use a Dremel freehand, and tried with both battery drill and standard drill press, I'd say that this method produces the best results by far. It was definitely worth the investment, as it was getting the proper PCB drills.

The board above still has the toner on it, but below you can see the under side of the control panel and how things fit together. The switch to the bottom left is our 'missile launcher' and the green board is the LCD.

Below is one half of the control panel all assembled. It is not apparent on this photo, but the cream coloured connectors at the top of each strip are there to drive the LEDs inside the push buttons. There should be three on each strip, but because the black IDC connectors have a larger footprint than the outline on Diptrace, there is no room for the closest of these connectors.

The missing connector has been soldered to the copper side of the board which is where the connecting cable below disappears to. It was the only oversight with the board, so I'm pretty happy with the overall result. On the photo below you can see that one of the button LEDs is lit. The idea is that the microcontroller will determine which LEDs light up - and if they stay on or flash or whatever. This provides a lot of flexibility in the design.

Below there are eight panel strips, four on each board. To simplify board manufacture and flexibility, they were designed to connect to each other via a ribbon cable. This means that a switch ladder can be built to almost any size and dimension. The HERMAN panel will have 20 switches, four banks of three (at left) and four banks of two (at right). You can see the connecting cable best between the two boards. The driving voltage (on the resistive divider) comes via the cable that disappears at top left, and the output voltage gets fed back into an analog input on the picaxe via the same cable. The first switch has a header link in place, and the last switch has a footer link in place.

The photo here did not show up so well, but here I am testing that all the switches are doing what they are meant to. So far, so good ...

A couple of minor circuit modifications were needed to make the control panel work as designed. The header resistor RO1 was halved in value, and an extra ladder resistor was put in before the first switch. In reality, this simply meant the first resistor (R1) was doubled in value.

At the same time as getting all the circuitry together, the control panel is finally coming together as planned over a year ago. Because we ran out of time, we been using this crude/hybrid/taped together ugly thing that has worked as a test panel, but only just.

So with the circuit boards happening, and the planning all sketched out in 3D, it was time to make it happen ... (with a deep breath that it would all come together in time for our next brew).

So this:

Is slowly being transformed into a real panel.

The panel is the old back panel that was ripped off to give access to the wiring cavities.

The version 5 interface board (shown above at left) is based on a picaxe 40X. The circuit board was our only attempt (so far) at a double-sided board. It was not well made, with etching and alignment issues. As far as design goes, it put too many eggs in one basket and was not easy to expand on. The 40X was the best available at the time, but the functionality of the X1 series almost makes them look prehistoric. Even though a 40X1 could be plugged straight into this board, it is near the end of its unreliable run. (The smaller board at right is a driver for the solid state relays).

The missile switch at top left is designed to work in reverse to the usual missile launch concept. We will use it as a quick shut-down of mains power in case it is ever needed. The LCD is a 40x2 Seiko unit. I had some trouble with initial testing because the datasheet is not overly clear but it needs a contrast voltage below zero to make it readable. It is an annoying bit of extra work - but I will source the minus voltage from the -12V supply on the PC supply that is powering the machine.

The paper with print is a drilling template for the pushbuttons, drawn in CorelDraw.

The pushbutton holes were drilled out with a forstner bit which gives a neat and accurate cut, especially compared to normal metal drill bits.

So all that is needed now is a bit of mounting hardware and the control panel strip PCBs to be finished and we will celebrate the result by making brew 2 of our double Dortmunder batch. I'm still waiting on some nice stainless fittings for the mash tun, but hopefully they won't be too far away.

It has been a good day. After what seems a painfully long time planning, we have finally been able to build some things. About a year ago we upgraded most of the hardware on HERMAN, an interim step towards version 6. This included a new brewing stand and new plumbing layout. We were very happy with the result - except that holiday leave is too short and the job was only half done.

We had big plans for upgraded control gear - to bring HERMAN into the 21st century ...

I've blogged before about HERMAN's control history. This time around we wanted a stand-alone micro processor controlled machine that could also be remotely controlled. Up until version 5 the machine could only be controlled via a laptop. This was disastrous whenever the machine crashed - which has probably happened a dozen times. We want this version to work without the laptop - and with the laptop - and with the internet as well.

Today we began to build the printed circuit boards for the control panel. We were using new techniques so there has been a lot of learning involved - both software and hardware.

Rather than hacksaw or bandsaw the board to size, this time we used a cutter to score the board and break it along the scored line. It seemed to work very well, especially by minimising high edges on the copper to be etched. It also made board clean-up a bit easier.

We also used cheap advertising paper rather than the blue 'press n peel' sheets. Apart from the difference in price, the most notable advantage is that dog hair is not attracted to the paper like it is to the blue sheets. It was hard keeping our earlier boards hair free!

We also used a laminator rather than a clothes iron. Apart from simplicity, it is much more predictable and therefore repeatable. We did ten passes through our particular machine and got consistent results. The whole process produced a much better board than we have been able to before - so I'm guessing it is because of a lot of small adjustments rather than one big thing.

I've never done a silk screen print on a board before. Partly this has been a software issue, but also when using the blue press n peel sheets it seems a bit of a luxury because the sheets are quite pricey.

We've played around with glossy photo sheets as suggested by US PCB hobbyists, but never got them to work. Here in Australia it is not easy to simply order a particular brand and code of paper from some overseas mega-mart.

A while ago I found an explanation about why we had so much trouble with paper. Essentially, the quality of glossy photo paper has improved a lot over the time inkjets have been around. Older clay-based papers have given way to newer water-proof ones. The problem lies in the waterproof coatings because newer papers will not dissolve away from the toner.

The suggested solution is to find cheap and nasty glossy paper - and it is the stuff that gets delivered seemingly daily to our doors - the paper under all that glossy junk mail advertising.

Rather than test it on ultra-clean copper intended for etching, it seemed a good idea to try a silk screen on an old board that we were never going to use.

The print on the paper is barely recognisable due to the glossy 'pre-print'.

After 10 passes through a hot laminating machine (the type used to laminate plastic covers on business cards and the like), the paper had stuck to the board substrate. It was now time to soak it in water to see if the paper would release and leave the toner intact.

And hey presto - the paper felt slimy the moment it hit the water and was easy to rub off, leaving a beautiful print on the board - except for the fact that I forgot to mirror the print

If I can get the toner to adhere that well to copper, I'll be very happy. Getting ever closer now to making some shiny new boards for the HERMAN project .

I realised that as the complexity of HERMAN was growing I needed to do some layout drawings just to make sure things worked in terms of both wiring and fit.

So I immediately turned to my old friend CorelDraw to do mockups of the mash tun and produced this in short time.

But I then realised that as it got more complex, I needed to do a top view to figure out if the liquor in manifold and the mixing paddle would actually work together.

It was at this point that I remembered why a 3D drawing might actually be useful. A while ago I discovered Google SketchUp and immediately became a fan of it. I know that Garrett has knocked up some nice models of his brew sculpture and various components such as ball valves and thermometers. Having worked partly through the learning curve on both VSM and DT, I was not in the mood for learning how to drive another complex program.

But with some spare time on Sunday night I tackled some of the tutorials. It seemed even easier than I remembered and I wish I'd checked out some of the videos when I first came across SketchUp. It is one seriously capable design program. For a person like me who feels 3D challenged, it is extraordinary.

One of my interests is woodturning. I can see how SketchUp can help with designs for this. No more graph paper for me.

But to my own brew sculpture. The challenge was to re-create the drawing from CorelDraw into a 3D one. After about 3 hours of learning, I got this far.

The basic mash tun was drawn in a few minutes. A few minutes later I had a nice straight edge on the top of the tun - like a chamfer. The rest of the time was spent working through the sequence of tutorials to figure out how to use the follow me tool and create a nice tun shape. I can see the power in this program - its just that while some things are incredibly intuitive, other things are not until you learn their particular way of doing it.

Anyway, there are so many advantages to drawing a brew sculpture in 3D, and the program is really amazing, so the learning curve is worth climbing.

After a bit  more work, the tun was refined a little, but still a long way short of being more useful than the 2D drawings from CorelDraw. I guess that a bit of persistence it will make a difference. I've always struggled to be patient ... must wait ... must learn ... must hold off to get it right this time ...

Back in the workshop on Monday, I planned for a gentle day pottering about looking at things with a critical eye and doing small improvements - like fixing the false bottom.

The open hole in the top sucks air in. The tee was originally plugged at the top. It is a tee rather than an elbow to provide a mechanical point for the bottom of the mixing paddle.

So a soldered plug on that offending end and a bit of silicon just for good measure seemed the trick. If it was being built from scratch the silicon wouldn't be needed, but it is difficult to solder such small pieces and not destroy existing solder joints.

A while ago I had a first attempt at some compression fitting disconnects that could be hand tightened by adding home built wingnuts to the fittings. They worked to a point but were hard on the fingers. I thought about a way to improve a promising idea and came up with this. It is definitely better to use, so time will tell if they are mechanically strong enough and how the fingers fare.

The bolts are 3mm with nylon spacers and a nut on the end (the nut is a simple spacer). The nylon with the rounded bolt heads are a nice combination. The holes in the compression fittings are tapped so the bolts screw in. The bolts are then strengthened with a dab of loctite on the threads.

Apart from some minor improvements and tidying on HERMAN, today yielded one more modification of note. The layout schematic shows that apart from the mash tun false bottom there are a couple of other filters in the brewing machine. One is in the kettle fill line to eliminate any particle matter (grains or husks) from entering the kettle. The other is the more conventional pickup from the kettle into the fermenter. For a long time we have zip-tied a hop sock onto the kettle fill line. Cleaning this has been a bit of a pain, and needed a new tie each time. I had some stainless mesh filter material on hand, so now we have a new filter on the kettle fill path.

Just for good measure, below is a photo of the filter at the bottom of the kettle. It uses the same mesh as above and forms a ring at the bottom circumference of the kettle. There is a slot cut into the bottom of this ring (with a nibbler) that is covered by the stainless mesh. This means that hop residue and trub has a difficult time getting into the fermenter.

The work on SketchUp and CorelDraw is important if we want to make everything fit. We are still waiting on some components that will finalise the layout of bits and pieces and still waiting on some tools that will help with PCB manufacture. Once I get my head around SketchUp and get those bits and then have time to deal with the detail of it all, there will be no stopping us as we build the remaining bits of HERMAN version 6.

Oh ... except that there is still a lot of coding that needs to happen. So many things, and so little time.

After about three weeks of preparing for an overseas holiday and then enjoying Malaysia and Singapore, it was nice to get back to some HERMAN work. Last brew day I encountered a disturbing problem. Despite best efforts to clean the inside of the heat exchanger coil, flaky dried wort found its way into the fermenter and clogged up the plate chiller.

The good news is that the Belgian Wit is fine despite the unwanted flavour addition. It also seems that the oxidising cleaner soak I gave the plate chiller over the last two weeks has eaten away any remaining flakes.

This morning I began to think about altering the plumbing to improve the flushing path through the chiller. This meant moving the 'hot liquor in' line close to the chiller, ensuring the high pressure hot water did not go via the heat exchange coil, bringing flaky debris with it.

I posted a revised layout in a recent blog but realised a problem with the layout. If a connection was made at the plate chiller, there would be no way of stopping flow into the kettle if the hot liquor solenoid was on. The schematic below fixes this issue. It shows the hot liquor line connecting on the pump side of the backflush valve. Each vessel can now be individually isolated via valves.

The schematic has been refined a little to make the plumbing layout clearer. The full picture is available as a downloadable jpg file, and the original corel draw file can be downloaded as well.

Next steps include building some shallow walls on the bottom shelf of the machine to hold stray water (as they say 'leak happens') and stop it running directly onto the laundry floor. I'm also intending to do some picaxe code tweaking to account for the change from RIMS to HERMS configuration. Beyond that a new switch panel for the control system and then final testing and building a picaxe 28x1 based control system. That will more than see my final week of leave out.

I put HERMAN through his paces yesterday just to double check things, calibrate temperature probes, sort out any leaks and to give him a cleaning run.

After sorting out a few drips here and there, I used my new digital temperature probe to check calibrations on the machine. First, the calibrating thermometer was checked. I tested it on some boiling water and ice slurry to give two known points. It read 100.5 degrees C and 0.4 degrees C respectively. This meant that assuming it was linear in response, I would need to subtract around half a degree from its reading to get a calibrated reading.

I had done careful calibrations on the mash tun probe a few years back, and it was still within 0.1 degrees of calibration. I was happy with this result - I don't think I can realistically do better than that. The hot liquor tun and recirculating liquor probes both needed some adjusting. Adjustments are done within the picaxe in code - more or less a set and forget method.

As an interesting side observation, the calibration for mash liquor in (recirculating liquor) that I did yesterday afternoon was 1.7 degrees out this morning. The main difference between these two moments was the ambient temperature. Yesterday was a pleasant 21 C outside and in the afternoon the laundry gets the sun. Added to this, HERMAN had been producing heat for much of the testing phase. This morning was much cooler. I suspected this might be the case - and will try and insulate the pipe where the probe sits. I intend to insulate much of the plumbing, and maybe this is the time to finish that job.

The new HERMS coil was then tested with a white vinegar and water mix. There are a number of things I was interested in finding out.

• what is the temperature differential between a stable HLT and a stable mash liquor?
• how long does it take to become stable?
• is there a way to speed up transitions without overshooting?
• is the heating in the HLT sufficient?
• how does the HERMS coil affect liquor flow?

The temperature differential between HLT and mash seems to be about 3 degrees C. This is what I remember from previous experiments, and I guess will be improved by adding insulation on plumbing. The greatest advantage of insulation, however, might be more efficient temperature stepping.

The system is slow to change in temperature. I had forgotten how slow HERMS is compared to RIMS. It took about an hour to rise from 34C overnight to settle at 50C in the mash tun. I'd anticipate that with grains this would take longer. Today I will be brewing with the machine. I intend to experiment with pre-heating the HLT (without recirculating the mash liquor) and then "sucking" the extra heat out of the HLT for a quicker step in the mash.

The HLT is not heating all the time while the mash liquor is heating up. This means that the heating capacity is not being limited by HLT - rather efficiency of HERMS coil and liquor flow.

The HERMS coil has a significant impact on liquor flow. The tap can be fully open, but the resistance of the flow path means that liquor flow does not seem overly fast. This is not a problem, as if anything it is still a little more than my normal recirculation rate. Too fast a flow will compact the grain bed. It might actually be an advantage to have the coil naturally restricting flow to prevent compaction.

At this point it concerns me a little how slow "step" changes will likely be. I will keep thinking about HLT pre-heating as an option, and also if it is useful to have a heat chamber as a booster element. My inclination would be to increase the efficiency of the HERMS coil and put all the heating in the HLT. I will just have to make beer to test the theory!

I've had a very satisfying day tidying up loose ends. I've added things like a clip to hold the 'fermenter in' hose - rather than let it dangle. I guess I've finally done all those 'roundtuit' kind of things that have been nagging. Usually the machine only comes out of its hangar when we want to brew. Those days are busy enough without having to worry about maintenance or improvements.

The hot liquor tun pickup tube never reached anywhere near the bottom of the tun. We would commonly tip the tun up to allow the last few litres of sparge into the mash tun. A simple fix - the pickup should deal with all but a few millilitres now.

For a while I've thought that the M15 compression fittings that are standard fare in Australia would be even more useful if they could be finger tightened and released. Here is my first attempt at creating a wingnut from a standard fitting. The 15C nut is drilled and tapped and the bolts at 180 degrees are fixed with locktite. I am planning to add some project box 'feet' to make it less stressful on fingers. So far this is looking promising - I'll have to wait for a leak test tomorrow to see if it really works.

Below you can see one of these in place on the mash tun manifold (in) line. One of those little bits I got around to today was labelling taps, as you can see in the photo.

I also did the work to convert HERMAN from a RIMS into a HERMS. The old copper heat exchange coil was badly oxidised and had the odd green patch on it. It was cleaned up with elbow grease and a white vinegar mix.

The coil below is ready to insert into the hot liquor tun. It still needs a vinegar wash through the inside of it. That will have to wait until tomorrow. The beer in the photo is the last pour of my American wheat. It had a touch of rye grain in it and was delicious. Even sadder than finishing this keg was the reality that I now have no beer ready to drink . I can't remember the last time that happened. At least we found a fabulous local micro that does take-aways called 2 Brothers. The plan is to brew on Friday - and with a Belgian Wit nearly fermented out we will somehow manage.

Yesterday I was working on the wiring on the back panel. Duct tape makes sure that it is all held in place. Now all I need to do is build a new cover DOH

And HERMAN is now looking very neat. All the wiring is hidden. All the disconnects are functional without spanners. All the hoses now have a home. Even the pens and clipboard have places to hang. All I need to do now is install some basic shelving to hold things like water chemicals, nutrients, scales and the refractometer.

And with the upgrade on the control panel, HERMAN 6 will be complete.

I wheeled HERMAN out of his cupboard today intent on cleaning up a lot of loose ends. The typical pattern has been to patch him together as quickly as possible so that he can make beer again ...

It has resulted in some pretty ugly workarounds that it was time to fix.

One of the 'nearly made it' design features of the version 5.5 hardware was the back panel which contained cavities for running cables - and therefore space to hide them. The oversight was that the back cover of this panel was glued to the rest of it. First job was to rip this nicely painted panel off to expose the cabling duct space behind it.

When all the cabling is done I will make a replacement cover that screws on this time.

Below you can see the cavity space with some existing wiring for the 12 volt downlights.

The cabling from the control panel to various tuns was randomly draped across the front of the machine because access to this cavity was next to impossible. Next some holes were drilled to allow cables from the cavity to match up with the mash tun and kettle.

Above you can see the connection to the mash tun temperature probe. This is the other side of the back wall cavity.

Underneath the mash tun and kettle shelf there was a lot of wiring of 240VAC through the solid state relay box (the grey box on the right). These blue cables just draped down in a messy way and it was time to clean them up. I decided that some wooden pieces running from front to back would do for a holding space.

While it still looks a bit messy from this angle, the wires no longer hang down and are not visible unless you really stoop down to investigate. When all is done, a false bottom will cover the wiring much like the back panel cover.

Some time was taken to fix a few things with the mash tun. The mixing paddle shaft was renewed which meant soldering a key for the mixer motor. A new rubber mat was installed under the false bottom and a hose clamp added to make the pickup air tight.

I found the old copper coil used for the heat exchanger. I intend to clean this up and fit it into the hot liquor tun tomorrow. I should be able to brew on Friday and test out the theory then. I am hoping the hot liquor tun will not be underpowered with its 3600 watts of electric heating.

The poor finishing of cables has been annoying me for a long time. It was good therapy to make inroads on this today.

It just seems wrong to take a nice stainless steel urn and drill a hole in the bottom of it. There is an old adage amongst carpentry to measure twice and cut once, so we made sure the hole was going to be in the right place ...

The hole is actually the drain point from the kettle into the pump for dough-in and recirculation - this is the path from the kettle into the mash tun which will sit above the kettle. We want the kettle to sit neatly on top of a pressure fitting, and simply connect to the pump by putting it in place. For this to work, things have to be solid and in the right place.

Once the top board of the control box was cut to size, we worked out exactly where we wanted the kettle to be in relation to the pump. After some checking and double-checking, the hole was drilled and filed out, and the fitting installed. The photo shows the under side of the kettle. with fitting coupled to its mate.

We worked out the right height for the connector below the kettle, and added a block of timber to keep it solidly in place.

The photo below shows the two nearly coming together. The 'O' ring on the top connector fits snugly into the bottom connector and the kettle weight holds it in place.

We have yet to test it, but we're confident it will work ok without leaking. The connectors are standard brass garden hose fittings available in Australia. The locking part was removed from the bottom connector by removing a circlip.

The next few things on our hardware do list are:

• fit the pressure sensor into the bottom of the kettle
• connect the kettle to the pump inlet
• install mash tun supports and hang the mash tun on them
• connect the pump outlet to the line running up the kettle to the mash tun

The next few things on our electronics do list are:

• finalise the controller schematic, etch and build the board
• finalise the mash tun connector board, etch and build it

And on the coding list:

• code some simple manual control routines to get the machine running

With a couple of brews safely fermenting away in the background, we've had some time recently to return to the HERMLET project.

A lot has happened in the odd spare minute over the last few weeks. I have been going over the notes that sparked the machine idea and weeding out some of the complexity (as per the Simplicity Cycle). Some coding pressure has helped speed up this review, because I needed to work out detailed control procedures - which means making decisions about which features the machine would end up with.

Having gotten that far, it was time to make something.

We re-gathered the various bits and laid them out on a portable work horse to see things in 3D. From there some of the parts were put into place, like the float switch as shown below. This switch will make sure the heater cannot operate unless it is covered with liquid.

We then made a plan to visit the local Trash 'n' Treasure market to look for a suitable stand for the machine. It needed to have wheels for portability and stand about 600 mm (2 feet) tall. Impatience got the better of us though, and we paid a visit to the local hardware warehouse first.

There we discovered a suitable tool chest with wheels and great storage for all things brewing. The chest will be home to brew water chemicals, yeast nutrient and tools like our digital scales and refractometer.

Once this was assembled, it was back to imagining where all the bits might be...

The photo above shows some spacers that will form a frame for the control gear. The photo below shows a rough location of the power supply on the left and the pump towards the rear. The plastic tray underneath the pump will be our simple leak detector.

The leak tray is there for a good reason (it is not because we don't think we can put in plumbing that does not leak ;-). One of the key design features is to make the machine easy for Leah to use. She has restricted use of one arm so we will be installing an easy connect/disconnect system for the kettle. The kettle stands immediately above the leak tray as per the photo below. Because the pump and recirculating path is low pressure, we think we will be able to use simple brass garden hose connectors (see bottom photo) as press fittings between kettle and pump. 

This means that the connection will hold through the weight of the kettle and an 'O' ring only. A transparent leak tray will provide an early warning in case things go awry.

We had time to get the electronics gear out again last weekend, so we put together a test circuit for our level sensor.

Leah built our level test rig. The vertical is a hose with a funnel at the top to allow easy filling of water. To her left is the pressure connection to a motorola MPX5010GS sensor. This unit outputs around 5 volts at 50 cm of head pressure. For our purposes we will only need about 20 cm of this range, but the picaxe which will be doing the analog to digital conversion will still see around 100 mls of resolution. We figure there is no need to amplify the output (ie. keep it simple).

On the right hand side of the rig is a tap to let out water and therefore easily vary head pressure for testing purposes. (Thanks Garrett for the inspiration for this test rig ;-)

The circuit we put together has an LM3914 LED driver driving a LED bar display. This will give us a simple visual indicator of 10 LEDs to show a full tun. You can see our prototyping board above and the actual circuit diagram is shown below (sorry about the poor image here - this is a function of uploading to live spaces - please contact me if you want a copy you can read).

We marked out approximate volumes on the test rig, correlating head pressure (height) from the brewing tun onto the hose. The values shown are in litres.

 And the circuit worked beautifully. The indicator below shows the 'tun' is half full. Now that we have a proof of concept, we can design and build up a printed circuit board.