Bus Conversions dot Com Bulletin Board
October 30, 2014, 11:54:37 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
News: New ownership began September 1st 2012!  Please send any comments to info@busconversions.com
   Home   Help Forum Rules Search Calendar Login Register BCM Home Page Contact BCM  
Pages: 1 2 [All]   Go Down
  Print  
Author Topic: variable speed control for 12 volt fan  (Read 14024 times)
JohnEd
Hero Member
*****
Offline Offline

Posts: 4571




Ignore
« on: January 13, 2008, 02:02:16 PM »

Guys,

I use a auto radiator cooling fan for ventalation and cooking exhaust.  At 12v it ROARS but it evacuates the coach in a New York minute.  Close the door and your ears pop.  I have an ignition dropping resistor in series with it that is switched but it is either slow or super sonic.  I shy away from a reostat because it consumes as much power as the max draw all the time.  The advent of the SCR allowed speed control with a small, cool running device in 115 v circuits.  Are these available for 12 and 24 volt systems?  I have had this thing installed since 1990 and when I am living in the coach it runs slow for 24/7 and at higher speeds during some of the day.  Real happy with the longivity.  Speed controls for 15 amp devices???  Anybody!

Thanks,

John
Logged

"An uneducated vote is a treasonous act more damaging than any treachery of the battlefield.
The price of apathy towards public affairs is to be ruled by evil men." Plato
“We can easily forgive a child who is afraid of the dark; the real tragedy of life is when men are afraid of the light.”
—Pla
DrivingMissLazy
Hero Member
*****
Offline Offline

Posts: 2634




Ignore
« Reply #1 on: January 13, 2008, 03:15:55 PM »

Guys,

I use a auto radiator cooling fan for ventalation and cooking exhaust.  At 12v it ROARS but it evacuates the coach in a New York minute.  Close the door and your ears pop.  I have an ignition dropping resistor in series with it that is switched but it is either slow or super sonic.  I shy away from a reostat because it consumes as much power as the max draw all the time.  The advent of the SCR allowed speed control with a small, cool running device in 115 v circuits.  Are these available for 12 and 24 volt systems?  I have had this thing installed since 1990 and when I am living in the coach it runs slow for 24/7 and at higher speeds during some of the day.  Real happy with the longivity.  Speed controls for 15 amp devices???  Anybody!

Thanks,

John
The ignition dropping resistor is the same as a rheostat which is just a variable dropping resistor. They work exactly the same except one is variable and one is fixed. NAPA should have a rheostat that will solve your problem. As I recall they are about 25 watts and 3-400 ohms.
Richard
Logged

Life should NOT be a journey to the grave with the intention of arriving safely in an attractive and well preserved body. But rather to skid in sideways, chocolate in one hand, a good Reisling in the other, body thoroughly used up, totally worn out and screaming:  WOO HOO, what a ride
Nick Badame Refrig/ACC
1989, MCI 102C3, 8V92T, HT740, 06' conversion FMCA# F-27317-S "Wife- 1969 Italian/German Style"
Global Moderator
Hero Member
*****
Offline Offline

Posts: 4874


Nick & Michelle Badame


WWW
« Reply #2 on: January 13, 2008, 03:30:09 PM »

Hi John,

I used theese dimmers in my bus to dimm my 12v puck lights in the ceiling. They can handle 70watts each.

They can easly do a 12v fan...

Sailor Sam's is one of my favorite places for bus lighting needs.
http://www.sailorsams.com/mall/dimmer_d1201.asp

Good Luck
Nick-
Logged

Whatever it takes!-GITIT DONE! 
Commercial Refrigeration- Ice machines- Heating & Air/ Atlantic Custom Coach Inc.
Master Mason- Cannon Lodge #104
https://www.facebook.com/atlanticcustomcoach
www.atlanticcustomcoach.com
boogiethecat
Hero Member
*****
Offline Offline

Posts: 634



WWW

Ignore
« Reply #3 on: January 13, 2008, 05:22:21 PM »

Type "dc pwm control" into ebay and you'll find exactly what you want.   PWM is similar to scr technology but works on DC, and doesn't get hot and waste energy like rheostats will.
Logged

1962 Crown
San Diego, Ca
JohnEd
Hero Member
*****
Offline Offline

Posts: 4571




Ignore
« Reply #4 on: January 13, 2008, 07:37:40 PM »

Richard,

I have the dropping resistor installed and it lacks control.....either roar or coasts.  Thanks though.


Nick,

Thanks for that bgreat reference.  I will get those unless i can do better on ebay.


Boggie,

You are the man.  I didn't have a clue what i was shopping for in terms of a "term".  I'll try ebay right now.

Thanks

John
Logged

"An uneducated vote is a treasonous act more damaging than any treachery of the battlefield.
The price of apathy towards public affairs is to be ruled by evil men." Plato
“We can easily forgive a child who is afraid of the dark; the real tragedy of life is when men are afraid of the light.”
—Pla
Sojourner
Guest

« Reply #5 on: January 13, 2008, 07:44:14 PM »

Bottom line never use under rated DC speed control for your application. If your vent fan has amps rate on label…such as a example…5 amps at full speed then you need at least 2/3 or higher amperage rating variable control for starting current (5 by 3 x 2 + 5 = 8.3). You can add reversible DPDT switch with equal amperage or higher diode connected to both variable’s output and installed switch before motor.
DC speed control w/on-off:
http://www.zaneinc.com/user/Ds-ams-l.pdf

Or my choice that I have use for many years on electronic projects. It never fail as long as you use diode rated for your needs…double voltage & amperage rating to control power surge via switching & motor starting. It usually cost less:
http://www.cpemma.co.uk/sdiodes.html

All the above will work on dc light system but it non-peak-surging….meaning you can use diode rated at 10% higher whatever it required.
Logged
Tim Strommen
Electronics Geek
Sr. Member
****
Offline Offline

Posts: 303



WWW

Ignore
« Reply #6 on: January 13, 2008, 09:32:43 PM »

I agree with Boogie - "PWM" is the way to go.

PWM send full voltage pulses of 12 or 24 (what ever your system is at) but varies the "on-time" to the load to give it a net "current regulation" effect (if you imagine that a motor running at 12-volts 100% of the time is drawing 5Amps over time, a motor given 12-volts for 20% of the time would draw 1Amp over time).  This is the way most new devices work - as motor especially operate better when given their design voltage.  This is also how LEDs are being dimmed these days (since they need an appropriate voltage AND current to put out the right color/ammount of light).

Most PWM devices these days are using MOSFETs which only have a dropping voltage of about .25-volts, which really gives merrit to the efficiency of these controls.

Cheers!

-Tim

P.S. Below is a simple Motor Controller concept (for those who want to do it themselves).  This is a single low-side MOSFET wich only needs a 5-12volt input signal to turn the load (in this case a motor) on or off.  By putting in a Pulse-Width-Modulation signal (either variable frequency or variable duty cycle) on the "pulse" line, you can vary the speed of the motor or dim a light.  I like low-side because the heat sink on most N-Channel MOSFETs ends up being electrical ground as well (preventing shorts). The source of said pulse can be anything from a 555-timer circuit, or a micro processor, or my favorite a saw-wave generator with an analog voltage from a potentiometer going into a comparator to generate a fixed-frequency/variable duty-cycle PWM signal (I can put up a schematic if enough requests come down...).

Oh - and remember to use a fuse (not shown in the drawing below!)-T
« Last Edit: January 14, 2008, 03:39:09 PM by Tim Strommen » Logged

Fremont, CA
1984 Gillig Phantom 40/102
DD 6V92TA (MUI, 275HP) - Allison HT740
Conversion Progress: 10% (9-years invested, 30 to go Smiley)
Stan
Hero Member
*****
Offline Offline

Posts: 973




Ignore
« Reply #7 on: January 14, 2008, 05:36:56 AM »

My experience with pwm controls on DC motors was RF noise so I just used a MOSfet with a pot in to vary the input signal. The other drawback to any solidstate method is you cannot get full speed. It is surprising how much speed that last volt (dropped across the device) increases.
Logged
JohnEd
Hero Member
*****
Offline Offline

Posts: 4571




Ignore
« Reply #8 on: January 14, 2008, 10:50:09 AM »

Well, remember that old...."Too soon old, too late smart"?  I am still on board with that one.  I went to ebay andlocated a few of those pwm's for sale/bid.  I found one that owuld handle 12 amps and was no bigger than a quarter and in "kit" form no less.  Lucky me and cheap.  Reading everything I could find about PWM's TODAY, and after winning the auction, I learned MORE.  It seems that dc motors don't work well with "high frequency switched dc" as the moror windings are basicallly coil inductors and tend to block higher(?) frequencies than DC.  Sojourners steer to ZANE INC or ZANEINC.COM sells DC speed controlers specially built for DC motor speed control.  The discussions I have read indicate that DC motors can be most effectively controlled with a switching freq of 400 to 800 cps.  My pwm is running 4.5KHrz and that is modest as these devices go.  It is specified as a DC motor controler but further research indicates that might be a really small motor.  Mine will always be usefull for controling interior lights so all is not lost.

Should my device fail, Sojourners recommendation will certainly work.  Thanks, Guy.

To all, thank you for you info and help.

John
Logged

"An uneducated vote is a treasonous act more damaging than any treachery of the battlefield.
The price of apathy towards public affairs is to be ruled by evil men." Plato
“We can easily forgive a child who is afraid of the dark; the real tragedy of life is when men are afraid of the light.”
—Pla
Tim Strommen
Electronics Geek
Sr. Member
****
Offline Offline

Posts: 303



WWW

Ignore
« Reply #9 on: January 14, 2008, 04:27:50 PM »

Stan,

EMI/RFI from PWM drives mostly comes from the "switch", the "line" and the "load".  With proper decoupling at the switch supply (like a cap between the +supply and -return path) and a choke on the supply line - a lot of potential EMI/RFI can be avoided.  For an application like PWM control on a motor things as simple as having a supply and return wire for the motor (not grounding the motor straight to the bus chassis), and twisting the wires will help a lot more too.  Then, placing a 1nF and 10nF ceramic "disc" capacitor right at the motor terminals (or the brushes if you can get to them) to be controlled to cut down any noise from the brushes (again, if applicable) will also help.

As for the PWM itself, a common way around a single-frequency radiation (especially at about 50% duty cycle when a PWM looks like a square-wave) - is to use "spread-spectrum" PWM, where the duty cycle is selected by the user, but the PWM frequency for each cycle is randomized to lessen a specific frequency peak.  This is where a microprocessor would show its strengths as a signal generator for a MOSFET control.

As for losses with a MOSFET (solid-state switch) my favorite N-Channel MOSFET, the International Rectifier "IRF3805" drops about 33mV with a 10-Amp load - which means you're more likely to suffer "significant voltage loss" from the transmission lines to the load than the "switch".


JohnEd,

Don't worry too much about an "experiment", we all make mistakes Wink.  Anything between 75-80kHz should be okay for a motor control.  If you have the ability, try to locate the control itself very close to the motor being controlled.  This will help to reduce the potential for emissions (EMI/RFI), and is a popular approach to robotic "stepper" controllers nowadays... (that is - motor integrated control, with remote management).

Cheers!

-Tim
« Last Edit: January 15, 2008, 03:05:52 PM by Tim Strommen » Logged

Fremont, CA
1984 Gillig Phantom 40/102
DD 6V92TA (MUI, 275HP) - Allison HT740
Conversion Progress: 10% (9-years invested, 30 to go Smiley)
Rick Brown
Jr. Member
**
Offline Offline

Posts: 61




Ignore
« Reply #10 on: January 14, 2008, 04:55:42 PM »

The PWM circuit Tim posted requires a reverse diode across the motor to absorb field collapse energy.  The posted circuit will work fine for a resistive load, but an inductive load stores energy in its field and when the current is interrupted the field collapses and induces a reverse voltage over the motor.  That voltage is felt by the transisor and field energy dissipated there.  Most transistors designed for this type of work can withstand this when switched occasionally, but 900 times a second will burn them up pretty quick.
-RickBrown in Reno, NV
Logged
DrivingMissLazy
Hero Member
*****
Offline Offline

Posts: 2634




Ignore
« Reply #11 on: January 14, 2008, 07:31:25 PM »

Gee, after reading all this, I think I will just go down to NAPA and buy a rheostat and suffer the ten watts or so of loss. If you can not afford to lose one amp of power then you have much more serious problems. LOL
Richard
Logged

Life should NOT be a journey to the grave with the intention of arriving safely in an attractive and well preserved body. But rather to skid in sideways, chocolate in one hand, a good Reisling in the other, body thoroughly used up, totally worn out and screaming:  WOO HOO, what a ride
Sojourner
Guest

« Reply #12 on: January 14, 2008, 09:27:36 PM »

Tim Strommen…I always look forward to your knowledgeable interest in electronic projects and that is a great asset to me and others on this board. However I believe there are some non- electronic bus nuts looking for simpler design that is easy to put together without doing the fine tuning to work best. I am not as much up to date & knowledgeable as you. I have also done many experimental circuit & adjustment till is tuned for the application. If they want to use what you suggest is fine….they will learn more & get the fun of doing it to say “I did it”….Great! I believe what I have suggested this diode-in-series fan speed changer is very effective & reliable for this application. I am sure you agree this simpler system will work on this large RV fan motor unless they want a variable control.

For those who want to use “series diode” voltage step down design and from 1/3 speed to full…read on.
This design will not cause sine wave noise but by motor only. Each silicon diode has voltage drop of .7 volts. So each stage of series diode will reduce another .7v and so on. There is no waste of energy and cooler then PWM which is variable but this is in stepping down voltage system. This is good enough for large vent fan.
If you can solder wire with rosin core @ 63% Tin & 37% Lead…solder.
Then you can build this. If not but want to…. http://www.teamnovak.com/tech_info/how_to/solder/index.html

Parts list for series diode voltage reducer:

12p Rotary Switch #105-14571 @.....................$4.76
11 pc. 15a 1200v Diode #747-DSEP12-12A @ $1.78 ea
1pc 20amp slow blow fuse   
1pc rectangular metal electrical box
1pc rectangular extension for above box.
1pc round or square metal cover plate.
The above electronic parts are list in this link: http://www.mouser.com/
If you can find better source at lowest price…Good for you…let us know if you will.

12 positions rotary switch will start at 7.7 volts lower than full voltage. Label 1 to 12 on cover. #1 is lowest speed to #12 at full speed.

This will work at any voltage up to diode limit in volts. Above list of diodes is 1200v. In other words 24v will work but needs to double diode in series which will be 1.4v between each drop stage.
Follow the “B” schematic only this will be 12 positions instead of 6 and much lower current system. Make “R1” a 20 amps fuse.
http://www.cpemma.co.uk/sdiodes.html

For those non-electronic geeks….read this in simple terms about diode: http://www.allaboutcircuits.com/vol_3/chpt_3/1.html

I am sure you may still have questions…feel free to post. If I can’t answer, I know Tom Strommen can…our electronic Geek!

FWIW

Sojourn for Christ, Jerry
Logged
JohnEd
Hero Member
*****
Offline Offline

Posts: 4571




Ignore
« Reply #13 on: January 14, 2008, 10:03:48 PM »

Jerry,

It never dawned on me that I could get a voltage drop across diodes without heat.  Great application/design.  I need a box and knob anyway.  I think I will build one of these for every fan in the coach.....7 each including the destratification squirrel.

Thank you very much.

Richard,

I thought like you do.  Whats an amp to get excited about.  Seems my ignition dropping ceramic resistor gets hot enough to have your skin stick to it if touched.  TOASTY!  I guess that fan draws 15 amps or so and running at slow speed 12.5 volts are being dropped across that 8 ohm resistor.  I never measured it of course but if temps are any indication...well!  For Bonnie work it was a waste of power.  That and the problem with my liking to fondle "cool" resistors.  Thanks for your help.  Your comments and logic are good as far as I can see.

John
Logged

"An uneducated vote is a treasonous act more damaging than any treachery of the battlefield.
The price of apathy towards public affairs is to be ruled by evil men." Plato
“We can easily forgive a child who is afraid of the dark; the real tragedy of life is when men are afraid of the light.”
—Pla
Rick Brown
Jr. Member
**
Offline Offline

Posts: 61




Ignore
« Reply #14 on: January 15, 2008, 07:57:57 AM »

*
It never dawned on me that I could get a voltage drop across diodes without heat
*
You cannot.  Assume your motor is drawing 1 Amp that flows through a .7 Volt drop over a diode.  Mr. Ohm once said Power = Amps * Volts.  That is .7 Watt per diode.
-RickBrown in Reno, NV
Logged
Sojourner
Guest

« Reply #15 on: January 15, 2008, 09:21:40 AM »

The bottom line about heat from diode....design heat load of whatever 8 or 15 or whatever amps will be at a given factory spec. limit. However if you go over ampere rating....in time it will smoke to puff. The lower ampere draws thorough a given ampere rated diode the cooler it will be.

FWIW

Sojourn for Christ, Jerry
Logged
boogiethecat
Hero Member
*****
Offline Offline

Posts: 634



WWW

Ignore
« Reply #16 on: January 15, 2008, 10:10:02 AM »

John,
Motors work fine with "high frequency" switching, at least when you're talking about 4.5 kilohertz.   Maybe a few hundred Khz or Mhz would be too high for a motor but under 50khz it's a no brainer.  Your control will work fine.
EMI really isn't an issue with this kind of circuit.  The ONLY down side to using low frequency PWM (anything in the audio range) for me is that if you still have good ears, you may get driven nuts by the whine the motor makes at the frequency the PWM is operating at.  400hz is tolerable but for me up in the 2-6khz range sounds like a really annoying whistle.  Above 15khz for most of us old farts won't even be heard.

Tim's circuit, yes it needs a "catch" diode inversely parallel to the motor or as Rick pointed out.  More importantly, it's only half the circuit and needs a PWM generator... trivial for us elektronikers but elusive for someone who isn't in to electronics to fill in the blanks. As an example of how a mosfet is wired it's almost correct (catch diode missing), but it's useless as a "real" schematic- it needs a lot more parts to function.

Diodes and a rotary switch?  It'll work but electronically speaking it's EXACTLY the same as a rheostat, just scads more complicated to build and gives you steps instead of a smooth transition between settings.  If you're even considering this, I'd suggest that you just get a proper rheostat (although I would never use either myself).  As Rick properly pointed out, It'll waste exactly as much heat at a given motor speed as will a rheostat, as the diodes are basically functioning as resistors in this application.
  I do have a place I love to use diodes with motors... with my drinking water pump.  I found that my sureflo pump went too fast for my drinking water spigot, and so it pulsed on and off quite annoyingly.  Putting two hefty diodes in series with the motor slowed it down just enough that the spigot was now faster than the motor could pump and the pulsing went away.  But that's the only kind of application I'd consider using diodes to slow down motors for...otherwise too many parts to do the job poorly that a decent PWM would do better.

Stan, I'm surprised at your comments "The other drawback to any solidstate method is you cannot get full speed."  These days as I'm sure you're aware, mosfet "on resistance" in the "Switched" mode (ie how much resistance is measured across a mosfet when fully turned on) is down in the milli-ohms. It's totally common to see less drop across a fully on mosfet than the average wire hooking to it.  Full speed with solid state circuitry is an everyday occurance in cars, battery powered tools, and basically everywhere.  The way you used your mosfet was probably in the "follower" mode (a MOSfet with a pot in to vary the input signal) and in that case you'll never get fully "on", instead you'll only get your power supply voltage minus what it takes to turn on the mosfet, usually 2-3 volts.  But this isn't a good way to do motor speed control, because now you're using the mosfet in "linear" mode and just like the switched diode thing, your mosfet is simply acting like a variable resistor and wasting exactly the same energy as would a simple rheostat.


Anyway, the bottom line is that the best way to do this is to use PWM.  For what it's worth, that's what is used in every battery operated variable speed tool... because it's simple, efficient, cheap, small, and works the best.  Maybe a suggestion would be just rob the PWM circuit out of a 14 volt drill, and there you are....

Cheers

« Last Edit: January 15, 2008, 10:13:21 AM by boogiethecat » Logged

1962 Crown
San Diego, Ca
Rick Brown
Jr. Member
**
Offline Offline

Posts: 61




Ignore
« Reply #17 on: January 23, 2008, 05:59:55 PM »

Here is the PWM circuit I use to drive a 24 Volt 16 Amp blower motor -> http://home.att.net/~intermountainac/PWM_1.jpg

As one never knows what evil lurks beyond his circuit board the optical isolators (U18 etc) are a good bet here.  The signals C10 etc. may come from a source as simple as a 555 timer.  Note that the electronics side is a 3.3 Volt system so your resistor values may differ.  The 4 pin connectors attached to the transistors rotate CCW when mating with the isolator connectors.  The output is configured so that the electronics may control both high and low side drivers.  The IRF4905L transistors are P channel MOSFETS drawn as presented by a TO-220 package (a fabrication guide).  The 249K resistors are for keeping the transistors off when you remove the electronic control circuit.  Because we really switch the power to the load rather than the ground to a hot load this is a high side switch.  The lower right circuit shows the 'catch' diode (drawn as a TO-220 package) added for an inductive load.  Notice the diode is directly across a pair of wires that go directly to the load.  Local grounding as is common on our vehicles is poor practice for PWM circuits.  You should bring both the power and ground wires for your load directly to the diode. 

How fast to run it?  The transistor only dissipates power when it is switching (to a first approximation).  Measured switching time for the circuit above is about 4 micro seconds.  If, for simplicity, one assumes half of the full current and voltage exist during the switching time then the power dissipated by the transistor during that interval is (16/2 Amps * 24/2 Volts) * 4 micro seconds = 384 micro Watts.  This happens twice per Hertz so if you switch at 1000 Hz the power lost is about .75 Watts.  That is independent of the power delivered to the load.  Switch at 10,000 Hz and you need to worry about heating issues.  I run at 960 Hz and can hear that tone when low power is being supplied to the load, but for a load such as a pump or fan to be useful it's going to make some noise also which quickly swamps out the PWM noise.  In summary: Crank up he frequency until you can't hear it, but keep your finger on the transistors.
-RickBrown in Reno
Logged
gus
Hero Member
*****
Offline Offline

Posts: 3527





Ignore
« Reply #18 on: January 24, 2008, 02:01:31 PM »

This has been a very interesting dialog on modern electronics to an old (vacumn tubes) Navy electronics type, but I'm not sure the question was ever answered?

I would be tempted to go with the simple but effective rheostat from NAPA!!
Logged

PD4107-152
PD4104-1274
Ash Flat, AR
JohnEd
Hero Member
*****
Offline Offline

Posts: 4571




Ignore
« Reply #19 on: January 24, 2008, 10:14:09 PM »

Boggie,

I have a PWM "kit" that i bought that has a switch in it numbered 1H46SB.  The description says that if you want to drive a device that takes more power than a small fan you have to add a heat sync.  I can get those but I am wondering what the max current would be with a proper sync.  Thanks

Rick,

I need a pwm that will handle 16 amps at 12 volts and the one you showed would do that....I guess.  I can't figure out the schem you provided.  i never heard of a optical isolator.  I am not clear on any of those drawings.  I can't tell what gets connected to what.  There are connections "C3 thru C8" and what goes there?  There is also a C3 thru C8 that appears to be terminal boards....connected to what?  Do the ckts on the left interconnect with the ckts on the right?  How?  I am really in the dark here and I was a tech for 30 years and that was tubes and transistors but i think I should understand something....and I don't.  Where is the component that varies the switching frequency?  I understood that caution you gave about keeping your finger on the transistor while increasing the freq.  Good advice!

Thanks and I hope you can give the guidance at the level of detail, read that as really dumbed down, I need to build one of these and implement it.

John
Logged

"An uneducated vote is a treasonous act more damaging than any treachery of the battlefield.
The price of apathy towards public affairs is to be ruled by evil men." Plato
“We can easily forgive a child who is afraid of the dark; the real tragedy of life is when men are afraid of the light.”
—Pla
Rick Brown
Jr. Member
**
Offline Offline

Posts: 61




Ignore
« Reply #20 on: January 25, 2008, 03:12:48 PM »

Here are some rheostat calculations.  Keep in mind that Watts = Volts * Amps and Volts = Amps * Ohms.  Assume a 24V, 16A load is resistive and you want to cut the power to it by a factor of 2 (24V*16A=384W & 384/2=192 Watts) with a rheostat.  The resistance of the load must be 1.5 Ohm (24V/16A=1.5 Ohm).  At half power through the load (from the basic equation P=I^2*R) the current is 11.3A (sqrt(192/1.5)=11.3) and the voltage across the load is 11.3A * 1.5 Ohm = 17V and the voltage over the rheostat is 24-17=7V.  The value of a series dropping resistor (or rheostat) must be 7V/11.3A=.6 Ohm and the power is 0.6*11.3*11.3 = 79Watts.  Big rheostat, but who cares as long as the engine is running and you have some place to mount it?

PWM becomes useful when you run from the battery.  I calculated in another post above a PWM circuit to control the same load would dissipate about 0.75W.

For the circuit: The optical isolator is a needless complication for a simple one-device controller.  Optical isolators consist of a light emitting diode (LED) and a photo transistor (PT) housed in a light tight package.  When you pass current through the LED it shines on the PT and turns it on to conduct current on the load side.  Thus the load is controlled by light and a big spark coming from the load side will not disrupt the control electronics.  I'm using a micro-controller for control of the PWM among other things and so use optical isolation to protect the control electronics.

Notice the vertical rectangle with four connector pins represented below the text “IRF4905L”.  That is a connector at the end of three 22 gauge wires which lead to the transistor mounted on a heat sink.  It plugs into the electronics board such that the top wire connects to the junction of the opto-iso pin 8 and the 1.00K resistor.  The unused pin comes into play when the circuit is a low side driver.

Sorry about the dual “C8 etc” notation.  C8 on the right is a connector number screened on the board.  C8 on the left is the name of a signal from another sheet.  Poor documentation practice: My excuse is I didn't think anybody else would ever look at it.  The larger circles with nearby “16ga” notation represent terminals on a barrier block interfacing to the outside world.

In my case a computer determines the switching frequency.  In a more practical, single-device, controller a NE555 timer like device would be the controller.  I see you can run a NE555 at up to 16 Volts so I suspect you could build a 12 Volt controller using only the NE555, it's timing components and the IRF4905L.

You can google up “NE555” or any of these device numbers and get the manufacturer's data sheets which tell how to use the product.

The TO-220 package is well suited for mounting to things metal which act as a heat sink.  Digikey part #345-1000 is thermally conducting, electrically isolating pads and part #4724K is mounting hardware.
-RickBrown in Reno, NV
Logged
Pages: 1 2 [All]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.18 | SMF © 2013, Simple Machines Valid XHTML 1.0! Valid CSS!