Aristo Mikado joins the fleet

Greg Elmassian

Greg Elmassian

Registered
8 Mar 2014
4,242
641
San Diego
www.elmassian.com
Nice, the PolySwitch (it's not really called a polyfuse strangely) PolySwitch Resettable PTCs Devices - Littelfuse is a nice resettable device.

If you look carefully you can get ones that have tighter tolerances, but while for sudden shorts, you rate like a regular fuse (trips at 200%), at constant loads that don't spike a lot, they trip pretty colse to the rating.

The "thermocouple" is really a thermal switch, right? So on off should yield pretty much full voltage... are you going to use 20 ohms on the resistor? If so, max amps of about 0.74 would mean a 2 amp Polyswitch might bet technically better, I doubt you will get much of an inductive "kick" at this current and voltage, even though the resistor is indeed wirewound.

Nice idea, and the only question is what duty cycle will be necessary to keep it "warm".

An interesting solution to a common issue.

Greg
 
  • Like
Reactions: 1 user
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
Thanks Greg!
Sorry, yes polyswitch :)
Correct the thermal device is just a normally closed thermal switch glued to the outside wall of the metal tank.
Two holes will be drilled in the tank, silicone bungs inserted and then the heating element put inside with the legs pushed through the bung.
I have a few thermal switches to try, rated at 25, 30 and 35C. I'll start low and try to compare the switching temperature with the actual water temperature.

I had wondered about the duty cycle. It will be interesting to monitor it. I think it might pay to pop an LED into the circuit so the heating cycle can be observed (I'm not sure though, how one could be incorporated between the thermal switch output and ground... edit: PhilP sells prewired >18v LEDs which should do the trick!)

I've seen a pretty simple water level detection circuit made with a power mosfet and resistor, but it would mean two more holes and bungs for the 'sensors' and I'd guess the polyswitch is safer given there's a heating element invovled. its not as if all the water in the jacket is going to evaporate during a run.
 
Last edited:
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
Well I've taken a close look at the tank... oh my gosh. The space to get the heating element into is pretty tight and whilst I was hoping the outer tank shell might disassemble, it doesn't.

So it's going to need a 4.7 mm hole for the element insertion, into a space that is 6mm. Gulp 20201004_094345.jpg
 
Last edited:
PhilP

PhilP

G Scale, 7/8th's, Electronics
5 Jun 2013
26,774
2,770
Tamworth, Staffs.
Darren,
If the water-bath dries-up, then the resistor will just get hot. - The polyswitch will only trip if there is a fault.

You *might* be able to use two polyswitch devices? - The first close to the battery, as a general protection device.

The second, you would glue to the resistor (I would not bother removing the casing, of the resistor). - You would have to test this, but my theory is if there is water, then the heat will be transferred into said water. If the bath is 'dry' then the resistor (and the attached polyswitch) wold get hot enough for the polyswitch to trip.

A 'water-level detection device' would probably be a better option? :think:

PhilP.
 
  • Like
Reactions: 1 user
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
Darren,
If the water-bath dries-up, then the resistor will just get hot. - The polyswitch will only trip if there is a fault.

You *might* be able to use two polyswitch devices? - The first close to the battery, as a general protection device.

The second, you would glue to the resistor (I would not bother removing the casing, of the resistor). - You would have to test this, but my theory is if there is water, then the heat will be transferred into said water. If the bath is 'dry' then the resistor (and the attached polyswitch) wold get hot enough for the polyswitch to trip.

A 'water-level detection device' would probably be a better option? :think:

PhilP.
Thanks Mr P
I'll hold off on the whole 'is the tank dry?' circuitry for the time being and diligently check the water bath.

If I can get the thermostatic heating working first off, I'll treat that as step 1. I've ordered all the bits and a little tin box to test with. I've ordered a >18v led from your good self which I will place across the heating element to give a visual indication of the heating cycle.

Once set up, I'll post some video of it all in operation

NB I've dropped the battery specs to an 11.1v 8.7ah li-ion, on size and cost grounds
 
Last edited:
PhilP

PhilP

G Scale, 7/8th's, Electronics
5 Jun 2013
26,774
2,770
Tamworth, Staffs.
NB I've dropped the battery specs to an 11.1v 8.7ah li-io, on size and cost grounds
Do be careful!
Not sure of the actual model of battery you will be using, but even for testing... Before anything else... Fit the fuse! :):nod::nod::nod:

I would expect the battery to be able to supply 'several-C' as a burst of current. - Things can get very-hot, very-quickly. :(:sweating::eek:
NOTE:
A Polyswitch is a thermal fuse, where the internal 'substrate' melts to break the circuit.. If the Poly does 'trip', it too will be hot.

PhilP.
 
Greg Elmassian

Greg Elmassian

Registered
8 Mar 2014
4,242
641
San Diego
www.elmassian.com
Clearly the circuit should be designed to handle worst case current... that would be the current drawn with the resistor in open air.

Next you want to really see if you can tune the circuit that if the resistor becomes uncovered (no longer in liquid) than the overcurrent protection would trip there.

If you get that down, you will probably be fine... I would try to design the system so that the resistor sits in liquid even if low (thus horizontal), and put the temp sensor low on the "water jacket" too.

I like the idea of the LED, will help system tuning if nothing else.

Greg

p.s. the "tuning" of the circuit does clearly involve setting the "on" resistance to draw the design current, thus the danger of "a burst of current" really does not exist unless you have some failure. A fuse really won't help much other than something failing (still a good idea), your tougher challenge is tuning the circuit to maintain the temp, and to not be "fooled" by running out of water in the bath.
 
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
Here's an example of the simple low water circuit I had been considering for step 2.
20201004_191048.jpg
In this example circuit, water touching both probes would cause the LED to illuminate. When the water level drops sufficiently to uncover the 'high' probe, the LED goes out.

I've yet to work out how the circuit could be incorporated to control the supply voltage to the thermal switch
 
Last edited:
PhilP

PhilP

G Scale, 7/8th's, Electronics
5 Jun 2013
26,774
2,770
Tamworth, Staffs.
Well the tank is (I assume) metal, so that is one side of the circuit..
 
  • Wow
Reactions: 1 user
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
Thanks for making me feel like a bear of very little brain Phil! How did I overlook that simple fact? :)

So all I need do is attach one side to the brass tank and I could reappropriate the filler hole (bottom of the photo) to use an insulating bung with the 'high level' side of the circuit...

20201004_200543.jpg
 
PhilP

PhilP

G Scale, 7/8th's, Electronics
5 Jun 2013
26,774
2,770
Tamworth, Staffs.
Thanks for making me feel like a bear of very little brain Phil! How did I overlook that simple fact? :)

Sometimes, we are all just too close to a problem to see it clearly.. :nerd::nerd:
 
  • Like
Reactions: 1 user
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
Re: the water detection circuit. To my limited thinking, it seems as though I could put the start of the heating circuit (the input of the thermal switch) after the LED on the mosfet Drain leg.

So the combined circuits would be something like this 20201004_224743.jpg
 
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
Basking in the sunshine, ready for steaming
20201006_125922.jpg


20201006_125952.jpg

20201006_125858.jpg
 
PhilP

PhilP

G Scale, 7/8th's, Electronics
5 Jun 2013
26,774
2,770
Tamworth, Staffs.
Nice!

How come you get sun, and I have just had a downpour of biblical proportions? :(:rolleyes:
 
  • Haha
Reactions: 1 user
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
Ah...thats's because our torrential downpour was yesterday. When I was outside all day building a shed :D
 
Fred2179G

Fred2179G

Registered
20 Apr 2017
507
126
USA
That valve gear is certainly very weathered. Was that you or just nature?
 
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
That was my attempt at weathering the valveger :D
My first, which is why it's 'sub-optimal'

The Vallejo oil top-wash was quite effective but I think I started too light with a Tamiya silver basecoat which I then tried darkening down with an ink/isopropyl wash. It's perhaps ended up too light and patchy
 
DGE-Railroad

DGE-Railroad

Registered
26 Jan 2020
331
24
48
United Kingdom
Well I've had a go at testing a non-invasive version of a tank warmer.

Bench-testing seems successful. I thought I'd have a go at using a peltier module to heat the water, rather than drill holes and immerse an element.

I also found a nice little programmable thermostatic control module. So for about £10 I have something that will keep the water jacket at around 30C

Current draw at 10.8v (intended battery voltage) is a whisker over 3A.

Here is a video of the circuit cycling.

Peltier tank heating test

Waiting for the water to cool is a bit boring. If you watch the video, just skip towards the end.

Here is the actual installation, using battery power.
The thermistor is a perfect fit for the fill hole so I may use that rather than obtaining an external reading from the tank skin.

Battery heating of water jacket
 
Last edited: