Jump to content

Welcome to Ultimate Subaru Message Board, my lurker friend!

Welcome to Ultimate Subaru Message Board, an unparalleled Subaru community full of the greatest Subaru gurus and modders on the planet! We offer technical information and discussion about all things Subaru, the best and most popular all wheel drive vehicles ever created.

We offer all this information for free to everyone, even lurkers like you! All we ask in return is that you sign up and give back some of what you get out - without our awesome registered users none of this would be possible! Plus, you get way more great stuff as a member! Lurk to lose, participate to WIN*!
  • Say hello and join the conversation
  • Subscribe to topics and forums to get automatic updates
  • Get your own profile and make new friends
  • Classifieds with all sorts of Subaru goodies
  • Photo hosting in our gallery
  • Meet other cool people with cool cars
Seriously, what are you waiting for? Make your life more fulfilling and join today! You and your Subaru won't regret it, we guarantee** it.

* The joy of participation and being generally awesome constitutes winning
** Not an actual guarantee, but seriously, you probably won't regret it!

Serving the Subaru Community since May 18th, 1998!

Guest Message by DevFuse

- - - - -

DagNabbit! Heater blower is toast! (to be archived)

  • Please log in to reply
55 replies to this topic

#26 Guest_edrach_*

  • Guests

Posted 01 February 2002 - 11:01 AM

The original resistors measured 1.1 ohm, 0.9 ohm, and 0.7 ohm respectively for a total of 2.7 ohm. Aside from the cooling of the resistors in the blower airstream, they didn't block the airstream much. Don't forget that any large heatsinks and such will possibly block the airstream slightly. Good luck.

#27 Guest_Subaroid_*

  • Guests

Posted 01 February 2002 - 11:43 AM

Ok, so I tested with the two sealed resistors I bought today.

2.2 ohm 50W

1.5 ohm 50W

The 1.5 was closer to what I wanted, so maybe that and another, smaller one. The 2.2 was
too much. Fan barely moved any air.

Problem was, the darn thing got pretty hot, way too hot to touch. Well, so did teh last remaining original wire coil too.

I'm reluctant to put anything that hot inside my interior, so I'm going to ask the shop to order a higher wattage one for me. 50W was the biggest they had in stock. They can order larger ones up to 100W. Next phase will follow much later, as the resistors need to be ordered from UK..

The values Edrach posted seem to make sense. So I will need 1.1 and 0.9 ohm resistors, something like 75-100W to keep the temps down. Excess heat will only blow the resistor.
The smaller resistace should help too. (Less power to transform to heat, less heat to get rid of)


#28 Guest_Fuzzy Dice_*

Guest_Fuzzy Dice_*
  • Guests

Posted 01 February 2002 - 01:49 PM

Those values are very low, and it seems the power rating is quite high. I'm afraid its time to try something else...a better solution. I'm going to look into making my own DC motor speed control that will fit into the space provided. I haven't pulled out the existing resistor setup yet to see what's going on, but my plan, if simple enough (free time is sparce), is to build a transistor controlled device that bypassed the need for large wattage resistors. That is, all of the motor current will pass through a FET (type of transistor). By pulsing the FET on and off you can control the amount of current delievered to the blower thus controlling the speed. This is old technology called Pulse Width Modulation (PWM) and is a very common way of controlling motor speeds. The big advantage is that there is far less power wasted heating up resistors and you can tweak the speeds to your liking. The downside is that if your not careful in your design it could increase the possiblity of introducing a slight buzz into your stereo system (but that can be removed with a filter).

First step is find that resistor block and see which wires go where? So, its under the glove box someplace...?

Not this weekend...snowboarding...Mt. Baker.

#29 Guest_Subaroid_*

  • Guests

Posted 01 February 2002 - 02:02 PM

I might be ok with a larger solid resistor, it would still get warm, but hopefully less so.

I will also ask them if they have that special resistor wire I could buy, then i'd "roll my own" LOL

I mean make my own coiled wire resistor like the original.. hehe

I do have a brand new "general purpose" heater switch right here on the desk with me, and it has two of those coiled wire resistor elements. And the values match nicely. It would be SO easy to just cut them off and transplant onto the subary resistor block.

But that would not provide a "repeatable", easy solution for anyone else to benefit from. I'm not doing it just for myself. Believe it or not, I'm hoping to find a way to pay back to the group here for the great help and advice I have leeched off of you guys :-)

The new switch came from a local car parts chain, and cost about 50 bucks. Not a cheap solution either. Could not believe a small switch would cost so much. Bought it for my VW projects but have not used it yet. This switch has 3 speeds vs. the subaru 4 speed system.
(3 would be enough IMHO)


#30 Guest_edrach_*

  • Guests

Posted 01 February 2002 - 08:30 PM

For all the experimenters out there, I can save you a trip to the boneyard; I'd be glad to send you a bad one for the cost of shipping (typically $3.95 for Priority Mail). Or a good one for $13.95 (shipping included). I still think the best solution is to "roll you own" but a solid state device as described in an earlier post would work too; besides, it doesn't have to go into the same space so size shouldn't matter as much.

Reminds me of a story when I went (in the early '70's) to my local BMW dealer to get a directional flasher for my 2002. I was shocked when he charged $29 for it. I asked "Why so expensive? The last one I bought was only $1.98!" The reply was that the new improved version was a "solid-state transistorized" device which would last the life of the car. Given the life of my cars at that time was 10 years, I opted to buy 5 more of the $1.98 version over that time frame. I'm just too stubborn.

#31 Guest_maccee44_*

  • Guests

Posted 01 February 2002 - 11:00 PM

after skimming through the posts it seemed to me that none of you are having the same problem as me. My car is suddenly not having any heat come out of the vents, on all settings, however, i can hear the fan working and cold air comes out(its maybe a little warmer than normal air but still cold). Am I correct in saying that you guys had the problem of the fan going out and not the heat only?

#32 Guest_Bill Putney_*

Guest_Bill Putney_*
  • Guests

Posted 01 February 2002 - 11:14 PM

The best I was able to do was find my Excel calculations for the resistor values and wattages. Apparently I did not commit my Digi-Key part numbers to a stored file, but I know I did put the information in a post, which I guess is now lost to perpetuity. A real shame since a lot of work and time went into the results.

The calculations are based on Ed Rachner's measurements with a little fudge factor for some assumed contact resistance in the measurements (learned from experience).

The last three lines of data are for whatever safety factor you want to design in (50%, 75%, and 100%) for the wattage ratings. Actually the safety factors are much better than that, as the resistors are rated for wattage at 40° ambient air temperature without heat sinking rather than 25°C ambient - makes for a very conservative design.

Find the inexpensive square-body ceramic resistors in the DigiKey catalog (available on-line - www.digikey.com - but, as most things are, it is much easier to browse thru using the hard copy) and use series and parallel combinations to achieve the resistor values and wattage ratings (IIRC, the best choices happened to be parallel combinations of the available values; also all individual resistors may have been 5W or 10W IIRC). I believe going with 1.75 times the calculated wattage came out nicely to ganging several resistors of the same value to achieve total resistance value very close to those required and wattage more than 1.75 times those calculated (again, not including the fact that the resistor wattage ratings are for 40°C ambient).

If you need help with your calculations, several people here can give that. Basically, for parallel resistors of the same value, divide the individual resistor value used by the number of resistors. For series of the same value, multiply the individual resistor value used by the number of resistors. For wattage, multiply the individual resistor wattage used by number of resistors. (Those calculations *only* work when all resistors in a group are of the same value and wattage, and when all are either in parallel or all are in series - which, as I said, is how it conveniently worked out for this application with the available resistors from Digi-Key.)

As stated in an earlier post in this thread, you theoretically could use more compact, "prettier" resistors (with heat sinks) for a nicer package, but these ceramics are typically about $0.60US each vs. $3 to $5US each for the nice ones. Plus, the nicer expensive ones are only readily available in a *very* limited selection of lower resistance values that do not work out well for what's needed here.

THE REAL POOP (taken from Excel spread sheet):

Assumed measurement contact resistance (ohms): 0.05_ _ _ _
Assumed system voltage when measurements were made: 11.00 (Ed said he made his measurements with the engine off)
Assumed worst-case system voltage for wattage calculations: 14.00_ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _
Measured resistance values (ohms): RA = 1.10, RB = 0.90, RC = 0.70
Assumed actual resistance: RA = 1.05, RB = 0.85, RC = 0.65
_ _ _ _ _ _ _ _ _ _ _ _
Worst-case voltage drop as measured: VA = 3.10, VB = 3.00, VC = 2.90
Worst case v. drop at assumed worst case system voltage: VA = 3.95, VB = 3.82, VC = 3.69
_ _ _ _ _ _ _ _ _ _ _ _
Calculated wattage requirement: WA = 14.83, WB = 17.15, WC = 20.96
Req'd wattage @ 1.5 x calculated: WA = 22.24, WB = 25.73, WC = 31.44
Req'd wattage @ 1.75 x calculated: WA = 25.94, WA = 30.01, WC = 36.68
Req'd wattage @ 2 x calculated: WA = 29.65, WB = 34.30, WC = 41.92

#33 Guest_maccee44_*

  • Guests

Posted 01 February 2002 - 11:54 PM

gee, i sure hope that message wasnt to answer my question, because i have no clue at all what that stuff is.

#34 Guest_Bill Putney_*

Guest_Bill Putney_*
  • Guests

Posted 02 February 2002 - 12:01 AM

(see www.digikey.com, then click "Catalog", then "Catalog Sections", then "Section F - Resistors...", then scroll to page 9 of 57 for the square ceramic resistors)

For RA (the low speed resistors), use three 3.0 ohm 10W resistors in parallel (gives theoretical 1.0 ohm, conservative 30W; target was 1.05 ohms 26W).
Digi-Key P/N 3.0W-10-ND, qty. 3

For RB, use four 3.3 ohm 10W resistors in parallel (gives theoretical 0.825 ohms, conservative 40W; target was 0.85 ohms 30W).
Digi-Key P/N 3.3W-10-ND, qty. 4

For RC, use five 3.3 ohm 10W resistors in parallel (gives theoretical 0.66 ohms, conservative 50W; target was 0.65 ohms 37W).
Digi-Key P/N 3.3W-10-ND, qty. 5

(Note that RB and RC are built from the same resistor, so order 9 of those - actually you can get 10 cheaper than 9 because there's a price drop for 10.)

(BTW - I copied the text of these two posts into MS Word to save onto my hard drive this time.)

No, maccee44. This isn't to address your problem. This is to build a robust replacement for the factory resistor pack that controls the blower motor speed. You apparently don't have that problem since you said you blower works but you just don't get heat. Check that the damper door that controls air flow thru your heater core is working with the temperature control slider on the dash - there's a bell-crank gear under the dash that's operated by the slider cable that the screw sometimes comes loose on allowing the gears to come unmeshed - tighten that if loose - may have to experiment with the right mesh of the teeth of the two gears so that that door goes all the way from fully open to fully closed..

#35 Guest_maccee44_*

  • Guests

Posted 02 February 2002 - 12:12 AM

so is this for my question?

#36 Guest_Bill Putney_*

Guest_Bill Putney_*
  • Guests

Posted 02 February 2002 - 12:29 AM

See the very end of my last post - except what I addressed to you applies to my '86 - don't know how similar it is to the '82 for the damper door control. But it appears you either have a damper door control problem, air in your cooling system (which usually wouldn't just suddenly appear), or stopped up heater core (internally).

#37 Guest_maccee44_*

  • Guests

Posted 02 February 2002 - 02:03 AM

hey thanks, that seems like its it since i am having some problems with the controls anyways.
Just wondering if you added that end part adressed to me after I had posted, cuz I dont remember that being there and it makes me look kinda dumb :)

#38 Guest_Bill Putney_*

Guest_Bill Putney_*
  • Guests

Posted 02 February 2002 - 05:52 AM

No - you're losing your mind. Just kidding - yes I did go back and edit that in. I had to go back in to correct something else, so since it was so late and I was tired, I took the lazy way. The editing feature of this forum is nice to have, but it can cause a situation like that where one person is left feeling foolish - sorry about that.

#39 Guest_walkerscott_*

  • Guests

Posted 02 February 2002 - 10:02 PM

here is a link to a picture of the unit i just changed (bad unit)


#40 Guest_Skip_*

  • Guests

Posted 03 February 2002 - 10:48 AM

I thought the wiring diag from the FSM should be included
please see below link

#41 Guest_walkerscott_*

  • Guests

Posted 03 February 2002 - 11:08 AM

<img src=www.geocities.com/walkers...-117S.JPG>

<img src=www.geocities.com/walkers...-120S.JPG>

<img src=www.geocities.com/walkers...-125S.JPG>

#42 Guest_Bill Putney_*

Guest_Bill Putney_*
  • Guests

Posted 03 February 2002 - 11:53 AM

walkerscott - do the links just like you did on the first post, except start the URL with the "www." part - that is, leave out the "<a href="http://".">".</a>

Your second method works - click on the link but then manually delete the "greater than" sign on the URL line after it goes there - that's not the way it was meant to work on ezboard, but it will work.

#43 Guest_walkerscott_*

  • Guests

Posted 03 February 2002 - 12:20 PM

Posted Image
Posted Image
Posted Image

Hope this works


Hope this helps, Skip

#44 Guest_Bill Putney_*

Guest_Bill Putney_*
  • Guests

Posted 03 February 2002 - 04:25 PM

Hmmm. Didn't work for some reason. Oh well - people will just have to copy and paste the URL to their browser URL line.

#45 Guest_Subaroid_*

  • Guests

Posted 04 February 2002 - 06:26 AM

Ok, update follows.

I just bought a roll of resistor wire, 1.0mm diameter, 0.624ohms per one meter. Let's see what I can do with that. There were several different options, I would have liked to use the smaller size wire which would have had more resistance (less wire needed) but it would not handle high amperages. This is a compromise size, 4.22 Amps in 100 degrees Celsius or whatever. It can tolerate heat up to 300 celsius. Higher heat=higher Amp rating.

I have no idea if this will work or not, but I'll give it a try. Will report back later.


#46 Guest_edrach_*

  • Guests

Posted 04 February 2002 - 09:07 AM

Jan--lots of wire would be needed. The resistor coils in the unit are like 10 turns or less ( I can count one later ) and the would work out to about 7 or 8 inches maximum. Given the resistance readings (1.1, 0.9, and 0.7 ohms), you would need over one meter for each of the coils. That just won't fit in there. I've been told one can buy short lengths of various diameters of resistance wire at a hobby shop. I haven't found one here locally to try that out.

#47 Guest_Subaroid_*

  • Guests

Posted 04 February 2002 - 10:27 AM


First attempt gave pretty decent results.

I scanned a picture of the resistor pack I came up with.


The resistor coils had to be made that big, because I was playing it safe and used pretty heavy wire. If I made them any smaller, they would not reduce fan speed enough. The slowest speed is still faster than original.

I used 1mm thick wire, I should have used .8 at least. But I was unsure how much power the fan draws. Skinnier wire, like the one remaining stock coil resistor in the picture (3rd speed) glows dark red when the 3rd speed is engaged. The red goes away instantly once it is placed in the stock location, directly in the airflow. This makes me believe that it is safe to try the smaller diameter wire. This wire I used does get pretty warm, but nowhere near glowing red.

The advantage of going down on wire diameter, is also that it has more resistance. I can get by with less wire, therefore saving space. The way the coil pack is set up now, it does not fit in the stock location.

BTW, when I reached deeper into the stock square mounting hole, I found a very rusty, crusty metal "ledge" in there, surrounding the coil pack. I cut the opening larger with a knife, it was so rusty. The visible part of the mounting hole is plastic. I could have made the hole bigger so I could squeeze my monster resistors in there, but I think I'm going to go get a roll of that smaller diameter wire and try again. I want it smaller. Now I know how much heat to expect from the smaller resistor wire.

A roll of wire cost me USD 13 (must buy the whole roll, 12-14 feet of wire.. hmm.. I bet the extra will come in handy somewhere)


#48 Guest_Subaroid_*

  • Guests

Posted 04 February 2002 - 10:50 AM

Tripod doesn't allow direct picture links from another website, you need to cut and paste the URL, worx for me.



#49 Guest_Subaroid_*

  • Guests

Posted 04 February 2002 - 10:54 AM

Forgot to mention,

the original resistor wires are soldered to the terminals, but for some darn reason the wire I bought did not like the idea. Would not stick no matter what I did. Bummer. Ended up drilling little holes on the terminals, and used small screws to attach the wires. See the picture.

I could have tried welding, but thought it would be overkill and hard to undo.. :-)


#50 Guest_Bill Putney_*

Guest_Bill Putney_*
  • Guests

Posted 04 February 2002 - 12:35 PM

Yeah - Ed anticipated the soldering issue when we went thru this mental exercise a coupla months ago.

I don't know if silver solder and the right flux would work or not (maybe find a local expert welder and ask him). If silver solder would work, the other benefit would be a higher melting temperature than tin/lead solder. Otherwise, welding is the answer (in industry, they would do resistance welding - that's where you run a super-high current thru the joint for a brief period of time, as in milliseconds). If you can find a solder and flux to work, the best connection would be crimp-and-solder - the crimping gives the lowest connection resistance (base metal-to-base metal - good because solder is high resistance relative to base metal) and the solder gives additional structural support and shields from contact with air/corrosion. That type of connection is used all the time on production connections in the automotive industry.

Subaroid - you are discovering by experiment the formulas for resistance and wattage. Basically, fatter wire means lower per length resistance, so you need longer wire for the same resistance, but that buys you higher total wattage capability. So the trade-off (as you've learned intuitively) is power capability vs. physical size. The way you make it more compact with the same wattage capability is by use of higher temperature materials, heat sinking, and/or blowing air on it. Just thought I'd help you synthesize all the pieces of the puzzle that you've got sitting separately in your mind now. It's a wattage-density kind of thing. You get the formula for resistance and the one for wattage capability based on construction, then you solve them simultaneously to optimize the design.

To paraphrase a common saying in industry, you can have it cheap, you can have it compact, you can have it last a long time - any two of those three - but you can't have all three - that's the unifying equation of the resistor universe.

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users