Author Topic: Project Heat-Troller  (Read 10103 times)

0 Members and 1 Guest are viewing this topic.

rmorenojr

  • Guest
Project Heat-Troller
« on: November 07, 2009, 06:27:42 am »
Greetings, 

I'm attempting to build a custom heated grips controller for my motorcycle.  Since power usage is always a concern, switching a large power for low heat setting wouldn't be using any less power than when I'm switched to full heat.  So, I'd like to make a step-less heat controller.  I'd also like to be able to use a momentary pushbutton to turn on and off the circuit, and reset to off when I turn off the bike.  Unfortunately, an LED can be annoyingly bright at night, so I need the power LED indicator to automatically dim at night.  And lastly, I'd like it to indicate that the battery voltage is low.  Oh yeah, everything has to be as small as possible. ;D
 

rmorenojr

  • Guest
Re: Project Heat-Troller
« Reply #1 on: November 07, 2009, 06:30:27 am »
OK, here's my schematic: oh, my circuit drawing program didn't like labeling wire connector's with the same name.  So the top circuit wired connection correspond to the bottom circuit's with an "a" added to the label.  For example, "N" is wired to "Na".
« Last Edit: November 07, 2009, 06:58:15 am by rmorenojr »
 

Offline KuchateK

  • Regular Contributor
  • *
  • Posts: 78
  • Country: us
Re: Project Heat-Troller
« Reply #2 on: November 07, 2009, 06:42:46 am »
If you want many functions in small package why not use microcontroller?
 

rmorenojr

  • Guest
Re: Project Heat-Troller
« Reply #3 on: November 07, 2009, 06:47:51 am »
Some Explanation:
desired usage - 40watts total (20 watts per side)
I'd prefer not to use a relay, unless there's some small SMD relay, so I decided to use a quad Schmitt trigger use as a T Flip-Flop with a debouncing cap C1.  C3 and R3 are to soft reset the Flip-Flop at power up.

The dual op amp is used to compare the regulated voltage to the actual voltage (the trim pot R16 used to set the voltage value desired) and switch between the LED feeds.  The LED I was thinking about is actually two colors (Red/Green) within a single LED.

Lastly, the photocell (which is one I have in my junk pile) has a very high resistance when dark and very low resistance when bright light hits it.  With a little experimentation, I found that the resistor values R17 and R18 work well with the photocell to control the transistor Q4 that feed the green LED.

Any suggestions?
 

rmorenojr

  • Guest
Re: Project Heat-Troller
« Reply #4 on: November 07, 2009, 06:50:58 am »
If you want many functions in small package why not use microcontroller?

Would a microcontroller simplify the circuit design?  Any recommendations?
 

Offline KuchateK

  • Regular Contributor
  • *
  • Posts: 78
  • Country: us
Re: Project Heat-Troller
« Reply #5 on: November 07, 2009, 05:49:07 pm »
With microcontroller all "sensing" logic is built in. It allows to use PWM to regulate power to heating elements and LED brightness. For "switch" I would use MOSFET. Built in analog to digital converter would "read" photocell for light amount, voltage from battery and voltage (actually just on or off) to front light to determine "bike on/bike off" (depends on the country, in some bikes use lights 24/7). With atmel you can make something like this on ATtiny25. It has 2 PWM channels, 4 channel ADC. It is 8 pin DIP part. You can also add for example thermistors or LM35 to control temperature and have it fully automated :) For ease of development I would probably use Arduino and ATmega8. Chip is ~3 times more expensive, but Arduino libraries are allowing very rapid prototyping. So all parts you need are MCU, mosfet (or two for independent handle control :D), photocell, voltage regulator, led and some resistors for voltage dividers and decoupling caps :D Deluxe version has also temperature sensors. The rest would be in software.

BTW. I'm complete amateur, so don't take what I wrote to be good advice  ;)
« Last Edit: November 08, 2009, 04:21:40 am by KuchateK »
 

rmorenojr

  • Guest
Re: Project Heat-Troller
« Reply #6 on: November 08, 2009, 05:39:38 am »
Wow, the Arduino and the other MCUs look very neat.  However, those little boards are about twice as big and twice the cost of what I'm try to do.  Unfortunately, I didn't see any specification for temperature range.  Typical automotive temperature range is -40 to 85 or 125 degrees C.  However, if I ever plan to expand the features (auto temp control, etc.) I may have to take another look.  I went to the MakerShed web site and found all sorts of cool things.

As you can see from the schematic, two MOSFETs are driving the negative to the grip elements (controlled by the timer).  The other components are easily obtainable or I have laying around.

What I'm really concerned with is the way I'm trying to turn on and off the circuit using a quad Schmitt trigger.  I guess I'll have to purchase one and bread board it.

And if there are any other "Best practices" I should follow, please let me know.
 

Offline KuchateK

  • Regular Contributor
  • *
  • Posts: 78
  • Country: us
Re: Project Heat-Troller
« Reply #7 on: November 08, 2009, 05:36:58 pm »
Arduino is big, because it is prototyping board with programmer.

For final device you should definitely program standalone chip and there is only 28 pin DIP part to implement. There are extra parts on Arduino making it easier to use, harder to break and that increases cost. But you don't have to buy it. You just need ATmega8 chip, way to burn in boot loader on to it (very simple rs232/COM or LPT port adapter, you can find chips with bootloader already programmed), breadboard or some wires and RS232 cable to the computer (either USB or voltage level converter).

For simple circuits it is stupid to implement microcontroller, but after logic gets a little more complicated one microcontroller will cost the same as bunch of analog comparators, opamps or other logic components and you can implement very complicated routines that are very easy to modify.

http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1240978713

So for my first Arduino I paid about $3.60 for ATmega and $1 for MAX232 and maybe a dollar for few caps and oscillator. I plaid with AVR chips before, so I had ISP programmer. But Arduino is so much quicker to learn and do anything, that I switched from C to it and I got couple boards to play with.

Sorry for that off-topic.
« Last Edit: November 08, 2009, 05:43:01 pm by KuchateK »
 

Offline qno

  • Frequent Contributor
  • **
  • Posts: 422
  • Country: nl
Re: Project Heat-Troller
« Reply #8 on: November 12, 2009, 08:40:30 pm »
I am driving a BMW. This has a switch that switches the the heated grips in series or parallel. This is sufficient control for me.
What temperature are you driving in whenyou need the heated grips?
Why spend money I don't have on things I don't need to impress people I don't like?
 

rmorenojr

  • Guest
Re: Project Heat-Troller
« Reply #9 on: November 17, 2009, 05:34:52 am »
If I'm understanding correctly, the resistive elements within the grips are switched between series and parallel.  Not a bad idea.  So for example: if your hand grip elements are 8 ohms each, in series they would be 16 ohms and in parallel they would be 4 ohms.  Assuming 12.5 volts (battery), you'd consume 9.8 watts in series or 37.5 watts in parallel.  Much better than the heated grip manufacturer's recommendation of using a switched series load resistor which would continue to consume power regardless if switch high or low.

However, the stepless design will allow infinite variable controller (independently for each side) using a timer as PWM.  Varying the frequency would change the overall power consumed by the heated grips and thus the temperature.  In addition, I wanted the controller to reset off if I turn the bike off (or on).  Lastly, since motorcycles typically have an inadequate charging system, I wanted the controller to indicate that I'm exceeding the max amp draw.  I know a bit excessive, but a cool idea...right?

As for the temperature, if the roads are dry..I'll ride..even down to 20 degrees F.  Oh, I also have a full heated jacket liner.
 

rmorenojr

  • Guest
Re: Project Heat-Troller
« Reply #10 on: November 17, 2009, 05:38:54 am »
What I really need is someone to review my quad Schmitt trigger NAND logic, the soft/power on reset, and the momentary pushbutton circuit and let me know if it will work.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19491
  • Country: gb
  • 0999
Re: Project Heat-Troller
« Reply #11 on: November 17, 2009, 10:46:06 am »
You need to increase or decrease the duty cycle to alter the power level, altering the frequency does nothing.

You don't need an ADC for the photocell, a simple comparator will do unless you want the light/dark threshold to be programmable.

Here's how to use a comparator to make a light/dark sensor, replace the CdS cell with a thermistor to make it into a thermostat.
http://www.silicontronics.com/index.php?action=ezportal;sa=page;p=17

Attached is a circuit which will do PWM, Pot1 alters the frequency, in your case it doesn't need to be adjustable so replace it and R4 with a 47k fixed resistor, Pot2 varies the duty cycle which will alter the power level.

A latch can be made using a couple of transistors.

A quad comparator such as the LM339 could be used rather than several ICs.
 

Offline qno

  • Frequent Contributor
  • **
  • Posts: 422
  • Country: nl
Re: Project Heat-Troller
« Reply #12 on: November 18, 2009, 04:17:41 pm »
I usually leave the heated grips either in off or in the high position.
With your winter gloves it will take a while before you notice the heated grips.
And while driving in excess of 100 km/h the grips will have a hard time keeping your hands warm.
Specially when it rains.

I had trouble finding an enclosure to protect the circuits and components from the dust and rain.
I had build it in the fairing so it was sitting on the bike all year round.

I used a Bourns sealed potmeter.
I had a circuit that was a variation of the circuit above.

I once had a circuit published that had the same function.

http://notes.ump.edu.my/fkee/e-Magazine/Elektor%202004/Articles/E/e047031.pdf

Leave out R7. Replace R6 and C2 by a potentiometer as the circuit above.
When using a quad opamp you can control 3 heaters. I had my Jacket, and both feet on the heat control.

Why spend money I don't have on things I don't need to impress people I don't like?
 

rmorenojr

  • Guest
Re: Project Heat-Troller
« Reply #13 on: November 26, 2009, 05:43:28 am »
Quote
A latch can be made using a couple of transistors.

Hero999, could you sketch out this latch circuit?
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19491
  • Country: gb
  • 0999
Re: Project Heat-Troller
« Reply #14 on: November 27, 2009, 12:05:40 am »
Here it is.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf