Re: resistors

[a6561TB@gnn.com]

Mariott sez:
 
>according to
>>Igor Kessel
>>I highly doubt it. The stock resistor dissipates horrendous power. It is 
>>nothing but a pair of dumb ballast resistors , 0.2 ohm and 0.7ohm in 
>>series.
>>I wonder why Audi chose such a crude approach instead of using, say, 
>>thyristors. 

>your ballast resistor adds up to 0.9 ohms. this can easily be duplicated w/ 
>radio shack power resistors. they are about .25x.25x1.5 in.  using eleven 
>10 ohm resistors in parallel will give you 0.91 ohms with a power-handling 
>capacity of 110 watts. the resistors are ~$.79 for a two pack;  

Granted, I was talking about the wrong fan (my mistake), but let's see how 
actually incorrect am I...

Let's calculate the power we have to dissipate on the ballast resistor.
It consists of two resistors in series: R1=0.7ohm and R2=0.2ohm.

The fan has three stages of operation:

1st, slow speed: the current flows through both of the resistors in series   
                (total resistance Rt=R1+R2=0.9ohm) and through the fan.
2nd, medium speed: the current runs only through the Rt=R2=0.2ohm and        
                 through the fan.
3rd, high speed: Rt=0, the current runs directly through the fan.

Now we need to know the resistance value of the fan motor Rf. Direct 
measurment, as expected, was unreliable, Rf is of the same order of 
magnitude as the parasitic resistance of the leads and/or their contact 
points. Instead I measured the voltage drop on the fan, so I emulated stage 
No.2 by running the engine with the A/C on.

The voltage drop on the fan was Vf=6v. Since total voltage on the running 
engine Vt=14v, then the voltage drop on the resistor R2 is Vr=8v. Therefore 
the current flowing through R2 (and Rf) I=Vr/R2=40a! The power this poor 
sucker has to dissipate P=IČR2=320w. 
Now the fun part: to dissipate 320w on a pack of 10w resistors you have to 
use 32 resistors in a mixed parallel/series arrangement. Also this 
arrangement will have them runnng @ the max allowable temperature. Unless 
you want to be able to make yourself a nice over-easy omelette on the way to 
work in the morning, you might want to double the amount of resistors in 
order to halve the temperature they'll be generating.

The above considerations were for the R2 only. It can be easily shown that 
you also need about 20 resistors for R1 if you still want that "over-easy" 
or 40 if you don't. You are looking to install around 100 of them total to 
play it safe. But why would you want to use the Radio Shock resistors the 
first place? This is like trying to scratch a left ear with the right foot. 
It took me about 20 min to repair the OEM resistor (see my earlier post) and 
it's been working fine for more than 4 years now.  

You have also suggested that I

>if you have room for the resistors, try this setup to see if it can handle 
>the power, if not, let me know and i'll give you a different configuration 
>that makes 0.9 ohms and can handle it.

Well, I don't need to try it, I have this problem well sorted out a long 
time ago. This Radio Shock Resistor Simulation Theory is your brain child, 
so why don't you try it yourself.

Also check the Audi Service Bulletin 19-89-TO1:
It sez:   

Cause:   Fan relay stuck in "On" position.
	    Points welded together

Service: Use  -40amp relay, Second Stage
               P/N 443 951 253K  
              -70amp relay, Third Stage
               P/N 443 951 253J  

See, I'm telling you, that fan is a strong mother.
 
Igor Kessel,
New Product Development Engineer
MSEE, 1984