Vanatron: Camper in disguise! Let’s get IoT with a heat exchanger

;VENTILATION OR DEATH

One of the key issues in a campervan is removing water. Water being in lots of places that it shouldn’t be. No matter how well you insulate your van, there will be always be condensation surfaces. The vans themselves will always have little corners here and there that are poorly ventilated (mould hates ventilation).  Many things in a van add water.. c.900ml/day water is perspired/exhaled on average per person. So at peak occupancy over 8 hours, 2 people and one smallish collie will be adding well over a pint of water to a pretty small box.  Add to that wet clothes, washing up, cooking (for which ventilation is pretty important…)..

In general, the answer to this is more ventilation and massively overpowered heaters to compensate for the heat loss. This offends my sense of efficiency.

Enter, the heat exchanger!!  Theoretically, an air heat exchanger should perform a dual function in this use case. Remove water filled warm air from van, use it to heat incoming fresh drier air, save energy and fight mould. All this thing.

I found a chap had been working on a similar idea for a home made exchanger for a chilly flat. Someone else has done a lot of the work?? Get in.

Overengineering is for Winners

Whilst his design seems a good start for my needs, there are definitely some things that I want to add.  As this is going in the van, I need some wifi capability if I want to log the data nicely, so I’m using a Wemos Mini Pro as the base board for the project. This is a fantastic wee ESP8266 board that does wifi, talks arduino, and is basically made of pixie dust and wizardry.

wemos mini pro d1. Made of pixie farts. Trufax.

Firstly, I’ll have to make adapters for the air inputs, because I printed the parts before realising I have 100mm air intakes with 54mm on the exchanger. no big deal, but won’t be as lovely.

Secondly, I want to track humidity as well as temperature. For this purpose I purchased some DHT11 sensors.. and then promptly discovered that they are a bit rubbish, only do 1 degree granularity and apparently break in about two years of use max (I broke one already). I’ve ordered some BME280 sensors to swap in.

Note: The BME280 sensors arrived, and can only be set to one of two I2C addresses. This is problematic – without multiplexing them, I don’t have enough IO to talk to them over SPI.  I have compromised for the sake of expedience and ordered some DHT22 sensors. Slightly less crap than the DHT11s, they have the advantage of being a straight swap with one tiny code change.

DHT22. It looks just like DHT11 but is less terrible..

Third, I want to drive the fans automatically based on the sensor readings, and for this I’m using TIP120 Darlington Transistors to drive the 12v fans from the 3.3/5v board..

Fourth, I want some feedback. A related requirement is that I want some dimmer lighting for chilling with – the lighting I have currently is very efficient and bright and not terribly relaxing. So, I’m adding a small string of RGB LEDs to the board.

Pop on a couple of switches and we’re getting pretty maxed out on the IO. No matter, I think we’re done!

Lastly.. it’s not quite MaD sCiEnCe looking enough, is it? I think we can do better. I think we can.

The Build..

Assembling the main body of the heat exchanger was .. a little frustrating. After a few incidents – the last of which resulted in me staring into space and counting to a billion – I worked out a moderately efficient method using a bit of pole (and glueing the pox ridden ends on the tube. Most of a hackspace evening went on this, up to a gritted teeth finish at midnight when everyone was packing up around me..

It seemed like it was going ok.. Spoilers: it wasn’t.

 

*redacted* thing. Now I have a pokerer.

Ready to install, v1 code pretty much done!  The wemos takes data from the four sensors and posts it to an Influx database with a Grafana front end. Pretty, pretty graphs.

The code also controls a string of 6 LEDs, to provide a bit of calmer ambient light than the insanely bright overheads I installed.  The lights flash red (off), blue (auto) and green (on) to show the fan status.  Currently, auto just turns the fans on if the temperature or humidity in the van are high. At some point I will refine this to reflect the outside temperature etc.

Some refinement required
mad science lighting(tm)

There are a number of issues with the design that don’t really mesh with being in a van – the ends are not easily attached firmly to the tube, the air intake/exhaust ports are designed for small 6cm fans – and my ducts are 10cm. At some point I’d like to redesign so as to eliminate all the mad flexible ducting and jerry rigged plywood fan holders – but I want to have it running for a while to see how it performs as is. I still need to make a nice switch panel, sort out the internal air intakes and tidy the wiring a bit for this iteration, in any case.

Super Secret Science Project: WITH QUILTING

It started with some mud.

I’d been thinking about an alternative alphabet thing for a while, and when two scientist friends decided to make a new tiny scientist.. well. GAME. ON.

Playing in my shed with some clay, the idea to make an AMAZING nightlight out of porcelain arrived. It would look smooth on the outside but has the alphabet and pictures etched on the inside, so when the light is turned on – magic pictures!!

.. this idea lasted until I remembered my homebodged raku kiln will not fire to vitreous porcelain temperatures.

THEN I decided I would make fabric alphabet cubes, I would PRINT these. With PRINTING.

I purchased some lino and cutting tools. Awwww yissss toys. Then some more printing inks. More colours. IMPORTANT.

I started making letters (mostly) at random.

 

block printing for the win
block printing for the win

I went a little bit mad for a while..

han solo on everything
han solo on ALL THE THINGS

 

Then I decided that I would make the entire alphabet, and make it into a quilt.

Some maths later, I realised this would be about the size of a house.

The letters of the new person’s name! That is what I will do.  Well, balls. I have only made one of those letters so far.

I made the design..

draft design
Fibonacci, Bitchez

My fancy new phthalate, PVC, BPA, BBC, ITV free screen printing inks arrived..

painting on the design
painting on the design

 

finished blocks
finished blocks

There was a lot of fannying around decided what to use as a backing fabric. I ordered some bamboo velour (which is lovely), but then found some rust corduroy in my stash which seemed perfect. Warm and soft, yet tough and hard wearing. It contrasts very nicely with the coolness of the linen top, and provides a very substantial feel.

Next – fannying with whether to use quilt wadding (this is all new to me) and which wadding to use.

I want to make it be properly washable, so I was thinking of just having the linen top with a cotton or bamboo plush/velour backing, so that it’s hardly quilted at all. This is because I am suspicious of the washability of batting. Even if I am, I only want very light quiltingpaddinginserttechnicaltermhere.
Is batting bad for machine washing?
Actually I’ve got some polyester nubbly fleece and plain fleece (actually maybe i gave the last away I am not sure). Could I use that as the ‘quilting’?how about thin wool felt? That would survive a 40 if adequately quilted, right?

NEXT CONCERN
How do I make the layers stay together nicely for sewing in a way that doesn’t leave a residue? Can I use water soluble glue?
I think I have a walking foot.
I don’t know how it works.

I gleaned from the wonder that is Ravelry, that there are all sorts of strange glue sprays you can get (in america) and that they use elmer’s glue (which I worked out is pretty much just starch glue).

After endless research, I gave up on these crazy bought products and made some starch solution in a spray bottle (because a: couldn’t find this Sullivan brand mentioned by Intarweb ‘Merkins b: hate aerosols c: hate perfumed products) and used it to flatten everything out and then stick it together. It worked a treat! No pins! I hate pins. OH YES. YOU WILL DO MY BIDDING.

NEXT: a tester to check nothing would shrink in a horribly incompatible way and the ink wouldn’t come off etc etc etc

super pretty test thing
super pretty test thing made by vogons

 

Onwards!  Now I must finish all the block prints and their pictures..

OK this is not all of them but I am getting bored of all this typing.

The most troublesome bit was definitely the border. I watched a bunch of videos on youtube, whereupon I decided that a lot of people are bonkers. I played with some bits of paper and came up with the following method (which I did not entirely use, I made it much more complicated for myself)

making table mats with mitred corners
making table mats with mitred corners

Finally, I presented it to the new Autarch of Abingdon (HAIL)

She liked it a lot. I can tell.
She liked it a lot. I can tell.

TL;DR – I made a quilt!

A bad egg?

humans vs rainforestAn issue where consumer action can have a real impact on the way things work.. Unsustainable palm oil is hidden in many products, and is directly responsible for the destruction of millions of acres of rainforest. Think happy orangutan* thoughts and avoid Easter eggs from bad people!
See here for a table listing the results of a survey on palm oil use in chocolate.

*and countless other less photogenic species, indigenous humans, general ecological damage, greenhouse gas emissions and PUPPIES**

**not puppies. Except indirectly.