Hlt Controller

[matho]

just an update ran the controller yesterday all worked well might need a little tweaking to get everything right

cheer's matho

 

[matho]

ok it has been along time but i finally got around to stage two of the project. Volume measurement of the hlt
i got 4 load cells off a cheap electronic scale, they have two variable resistors in them one goes up with load and the other goes down. I wired them up into a wheatstone bridge arrangement, then put the output of the bridge into a high input impedance dc instrumentation amplifier. I have setup the amplifier so that full scale = 5v = 102.3kg, that should give me a 100 gram resolution when used with the 10 bit ADC on the PIC. I calibrated the amplifier today, with 25l it was low by 100ml so i think it is good enough.

ill get more schematics and pic up later.

cheer's matho

 

[matho]

ok here are some schematics



R1 adjusts the gain of the amp and R2 adjusts the supply voltage. R2 is needed because the amp can only supply an output voltage that is about 1.5v below the supply voltage so when i went to calibrate it, i first set the gain to full on the amp then adjusted the supply voltage to give an output voltage of 5v that would be full scale of the analog to digital converter. Then i adjusted the gain of the amp to give me a voltage rise of about 48.8 mV per litre of water.

R3,4,5,6 are there to represent the load cell, the 4 load cells are arranged into a wheatstone bridge so that two opposing sides go up with load and the other 2 opposing sides go down.




updated mainboard







some really bad pics

the whole sensor cost me about $30

so now i have to change the program to include the sensor.

cheers matho

 

[leiothrix]

Silly question - what are you using to draw your schematics?

Rob.

 

[matho]

Silly question - what are you using to draw your schematics?

Rob.

i use a program called cadsoft eagle its free with limitation of 50 components just google it


cheers matho

 

[RdeVjun]

Very interesting matho, exquisite use of cheap components and I'm certainly in awe of the ingenuity! :super:

FWIW, for work I just spent >$24K on 4 * CR1000s, 8 * AM16/32B MUXes (both CampbellSci) and PT load cells, can only resolve ~300g in a similar sort of thing. Slight difference is, I have 28 platforms with 84 cells, in groups of 3 for up to 1500kg! The data stream is pretty wide for that project but it is already delivering results...
BTW, looking to test some of these TI ADS1234 MUXes over the winter, 24bit and <$100, but it would be hard to justify (and integrate) in your setup, the 100g resolution you have already should be quite sufficient I would think. :icon_cheers:

 

[matho]

thats a nice little chip there RdeVjun, i was thinking of using microchips programable adc but as you said 10 bit resolution should be fine. I am worried about noise as the gain is above 200, might get it on my father in law's CRO to see the noise could need some fitering

cheer's matho

 

[MHB]

Totally OT

Mate is it Matho or Metho cos thats a Methanol in your avatar

Bad Alcohol

MHB

 

[matho]

Totally OT

Mate is it Matho or Metho cos that's a Methanol in your avatar

Bad Alcohol

MHB

i was wondering when someone would notice, just a joke on the chat that chappo continues with, i had 'meth' up as my avatar for a while but thought metho better.

cheer's
steve

 

[matho]

so today i got around to programming the pic and trying it out

here are some few pictures

the setup



with 10 l in the tun



with 15 l in the tun



i have a few things to fix up like how it zero's its self, i have already changed the program just have to reprogram the chip and try it out. It seems fairly accurate and linear so im happy with the result.
when i have tried out the changes ill post the code.

cheer's matho

 

[matho]

I have fixed a few bugs and are now very happy with how it works.
attached are the hex file and the pic basic program

View attachment Hlt_controller_2_Hex.txt

View attachment Hlt_controller_2_program.txt

and here is the final mainboard schematic



well thats it for the time being in the distant future ill be adding pid control but im in no hurry to do that.

thanks for reading and i hope this has inspired a few to do similar projects

cheer's matho

 

[QldKev]

i was wondering when someone would notice, just a joke on the chat that chappo continues with, i had 'meth' up as my avatar for a while but thought metho better.

cheer's
steve

and I thought it was your pom pom dog...


QldKev

 

[leiothrix]

thanks for reading and i hope this has inspired a few to do similar projects

Thanks for sharing :icon_cheers:

Rob.

 

[matho]

and I thought it was your pom pom dog...


QldKev

yeah it kinda looks like that doesn't :lol:

Thanks for sharing :icon_cheers:

Rob.

no worries rob i had fun doing the project and liked sharing it with others.

cheer's steve :icon_cheers:

 

[gap]

Great work Steve,

I will have to come down and see it in action soon.

regards

Graeme

 

[matho]

Great work Steve,

I will have to come down and see it in action soon.

regards

Graeme

that would be great graeme I'll let you know when I'll be brewing next no doubt it will be the first time I'll use the volume measurement in anger

cheer's. Steve

 

[MHB]

Ethanol is sometimes called the read headed dog, and water the Mickey Mouse Molecule, ah-well the things pissed chemists come up with
View attachment 38942View attachment 38943
MHB

 

[Zizzle]

ok here are some schematics

View attachment 38811

R1 adjusts the gain of the amp and R2 adjusts the supply voltage. R2 is needed because the amp can only supply an output voltage that is about 1.5v below the supply voltage so when i went to calibrate it, i first set the gain to full on the amp then adjusted the supply voltage to give an output voltage of 5v that would be full scale of the analog to digital converter. Then i adjusted the gain of the amp to give me a voltage rise of about 48.8 mV per litre of water.

R3,4,5,6 are there to represent the load cell, the 4 load cells are arranged into a wheatstone bridge so that two opposing sides go up with load and the other 2 opposing sides go down.

Only just catching up on this thread after matho posted in the arduino thread.

Very neat solution.

Just wanted to point out that you can get pretty cheap instrumentation amplifiers on ebay these days, like the AD620 or INA129 that simplify the circuitry. They can swing pretty close to the supply rails and can achieve gains of 1000 needed for load cells like this very easily. Just one external resistor needed to select the gain.

Stolen from this article:

http://www.ehow.com/how_5980113_make-load-...se-arduino.html

 

[matho]

Only just catching up on this thread after matho posted in the arduino thread.

Very neat solution.

Just wanted to point out that you can get pretty cheap instrumentation amplifiers on ebay these days, like the AD620 or INA129 that simplify the circuitry. They can swing pretty close to the supply rails and can achieve gains of 1000 needed for load cells like this very easily. Just one external resistor needed to select the gain.

Stolen from this article:

http://www.ehow.com/how_5980113_make-load-...se-arduino.html

ahh alot easier, i didn't even search for another solution i just went with a single supply instrumentation amplifier, Im really happy with how it is working it has made brewing more repeatable knowing my exact volumes in my HLT

cheers steve