What's new
CK5
Register an account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members.

Dual Circuit Windstar Fan - Final Shroud Update 05/11/11

Disconnecting the switch's grounds resulted in no change. It was worth a try though - Thanks for asking around. I'm not the type to drop it and "live with it" - There will be a solution in the end. I'll try a different power source next.


For those which are curious as to what 90k5blaze and I are talking about - Here ya go



There are a few ideas to try, hopefully prior to the weekend - Im wanting to install my water pump and noisey direct drive Saturday or Sunday.




Troy B
Ft Hood, TX
 
I am going to regret this......

OK, first of all, I am in awe of your fabricating skills and your patience.
You have created a truly fantastic setup.

However, you are way out in the bushes when it comes to the electronics.

I am an electronics engineer, with 30 some odd years of working on stuff like this.
I say this, not to brag, but assure you I know what I am talking about.

I am trying to not come off heavy handed, ask around, I get along with everybody. But you have several major flaws in your electricals.

First, your power blocks. Are those steel? They looked like it when you were building them.
Really really should have been made out of copper.
Silver would have been better, but would be totally overkill.

They can be saved though. Just make a thick copper strip to go along the top of the steel part.

Next, basic law of series DC circuits:
Current level is the same at all parts of the circuit.

ANYWHERE you break that circuit you are switching the exact same number of amps.
So ground switching is useless.

Next, any brush-type dc motor is also a dc generator. And vice-versa. I have seen old type generators squeal the belt if the points in the voltage regulator stuck when the motor was switched off.

I do not know of the fan motors in question are brush-type DC, or AC.
Given that they seem to be generating power when spun, I'm guessing DC.

And, before anyone jumps me, yes, they are going to AC motors in cars to a small extant. Not so much with the big motors like these.

Brush-type DC motors wear out and make electrical noise.
Plus they are inefficient.
With cheap solid-state equipment now, its easy to use a brushless AC motor on DC.
You just build an oscillator that converts the DC to AC.

All the 12 volt fans in modern computers are brushless AC.

As for the protection diode you first had in the circuit, all it would do would be to drop .7 volts going to the fan.
You need a protection diode across the coil reverse wired to stop what is know as back emf.
That occurs when the armature of the relay moves in the collapsing magnetic field of the coil.
Probably not needed in this case, and may be built in to the relay anyway depending on which model you have.

Now, the circuit problems you are having.......Dear God, whats with those switches??

You are showing a spdt switch.
It looks like you have the anode of the indicator diode on the common, along with terminal 85.
I assume that the switch is spdt, not spdt-center off.

When the switch is in one position, the anode and terminal 85 are hooked to ignition power.
But, when you switch it the other way, you ground those two terminals.....Why?

OK, just looked at that switch again, and I realized it must be a lighted switch. The way you show it is OK, but just a little misleading.
I am guessing now, its a spst. Not spdt. Makes a little more sense.

Remember, you do not know what is in those LEDs.
Since they run on direct 12 volt, and will compensate for voltage fluctuation, they do not have simple resistors inline.
As you discovered, that will not work.

When I build a 12 LED setup for someone, depending on who I am building it for, I may get creative.
They make LED driver chips that will take a wide variety of voltages in and output the correct voltage and currant for the LED you are using.

I often use one of those with a full wave diode bridge.
By using one of those, it does not matter which way they hook it up, the polarity will always be correct.

When you turn the switch off, of course, you are going to reverse bias the LED and blow it if its not protected.

The LED has to see a voltage DIFFERENCE across it in the correct direction to light up.
When the switch is on, but the coolant switch is off, the LED is seeing 12 volts on both sides of it.

Ignition switched 12v on one side, and 12 volt through the fan from the terminal block on the other.
What you have created here is the equivalent of a ground loop.
Remember I mentioned those terminal strips needed to be copper?

What is happening here is that something on the positive strip is drawing current, and there is enough resistance either in the block, or somewhere between it and the battery to cause a voltage drop.
So that the ignition 12 source is a volt or two higher than the terminal strip.
Which lights up the diode.

Or the fan rotating is generating voltage and back feeding the strip.

There are so many possibilities here, I'm not sure I can isolate the correct one without being there.
If you have a multimeter, and want to get to the bottom of this, let me know and I can probably talk you through some measurements that will solve it.

Its been a really long day, and I fell asleep twice typing this, so if it isn't clear, let me know and I will try to fix it tomorrow.

J.
 
Last edited:
Fordum said:
I am going to regret this......

OK, first of all, I am in awe of your fabricating skills and your patience.
You have created a truly fantastic setup.

However, you are way out in the bushes when it comes to the electronics.

I am an electronics engineer, with 30 some odd years of working on stuff like this.
I say this, not to brag, but assure you I know what I am talking about.

I am trying to not come off heavy handed, ask around, I get along with everybody. But you have several major flaws in your electricals.

First, your power blocks. Are those steel? They looked like it when you were building them.
Really really should have been made out of copper.
Silver would have been better, but would be totally overkill.

They can be saved though. Just make a thick copper strip to go along the top of the steel part.

Next, basic law of series DC circuits:
Current level is the same at all parts of the circuit.

ANYWHERE you break that circuit you are switching the exact same number of amps.
So ground switching is useless.

Next, any brush-type dc motor is also a dc generator. And vice-versa. I have seen old type generators squeal the belt if the points in the voltage regulator stuck when the motor was switched off.

I do not know of the fan motors in question are brush-type DC, or AC.
Given that they seem to be generating power when spun, I'm guessing DC.

And, before anyone jumps me, yes, they are going to AC motors in cars to a small extant. Not so much with the big motors like these.

Brush-type DC motors wear out and make electrical noise.
Plus they are inefficient.
With cheap solid-state equipment now, its easy to use a brushless AC motor on DC.
You just build an oscillator that converts the DC to AC.

All the 12 volt fans in modern computers are brushless AC.

As for the protection diode you first had in the circuit, all it would do would be to drop .7 volts going to the fan.
You need a protection diode across the coil reverse wired to stop what is know as back emf.
That occurs when the armature of the relay moves in the collapsing magnetic field of the coil.
Probably not needed in this case, and may be built in to the relay anyway depending on which model you have.

Now, the circuit problems you are having.......Dear God, whats with those switches??

You are showing a spdt switch.
It looks like you have the anode of the indicator diode on the common, along with terminal 85.
I assume that the switch is spdt, not spdt-center off.

When the switch is in one position, the anode and terminal 85 are hooked to ignition power.
But, when you switch it the other way, you ground those two terminals.....Why?

OK, just looked at that switch again, and I realized it must be a lighted switch. The way you show it is OK, but just a little misleading.
I am guessing now, its a spst. Not spdt. Makes a little more sense.

Remember, you do not know what is in those LEDs.
Since they run on direct 12 volt, and will compensate for voltage fluctuation, they do not have simple resistors inline.
As you discovered, that will not work.

When I build a 12 LED setup for someone, depending on who I am building it for, I may get creative.
They make LED driver chips that will take a wide variety of voltages in and output the correct voltage and currant for the LED you are using.

I often use one of those with a full wave diode bridge.
By using one of those, it does not matter which way they hook it up, the polarity will always be correct.

When you turn the switch off, of course, you are going to reverse bias the LED and blow it if its not protected.

The LED has to see a voltage DIFFERENCE across it in the correct direction to light up.
When the switch is on, but the coolant switch is off, the LED is seeing 12 volts on both sides of it.

Ignition switched 12v on one side, and 12 volt through the fan from the terminal block on the other.
What you have created here is the equivalent of a ground loop.
Remember I mentioned those terminal strips needed to be copper?

What is happening here is that something on the positive strip is drawing current, and there is enough resistance either in the block, or somewhere between it and the battery to cause a voltage drop.
So that the ignition 12 source is a volt or two higher than the terminal strip.
Which lights up the diode.

Or the fan rotating is generating voltage and back feeding the strip.

There are so many possibilities here, I'm not sure I can isolate the correct one without being there.
If you have a multimeter, and want to get to the bottom of this, let me know and I can probably talk you through some measurements that will solve it.

Its been a really long day, and I fell asleep twice typing this, so if it isn't clear, let me know and I will try to fix it tomorrow.

J.
Click to expand...

Glad someone brought up KCL after reading that stuff about running the relay on the ground side :D
 
You know its been 30 years, but I bet I could Thevenise that circuit too.

Its funny the things you remember.

When I was in Junior Collage, we had an electronics instructor named Robert C. Jones.
He always wrote his name as R. C. Jones.
Naturally we called him Time Constant Jones.

One day, he walked in and announced we were going to learn a concept known as Thevenizing a circuit.

He spent the next hour teaching us how to do it.
Everytime he would stop to see if we were getting it, we had it solid.

In fact, a couple of times we even jumped ahead a little.

At the end of the class, he announced that he wanted to congratulate us. He had been teaching that for years, and we were the best class he had ever taught. We picked that up so fast, we were just amazing.

Unfortunately, someone had to pop his bubble and tell him that we had all had that last year.
He had picked up the wrong course book when he left his office and thought we were the first year class.

When he asked why we did not say anything, we told him we thought he was going somewhere with that.
We all just assumed that he was reviewing before giving use something more advanced.

BTW, if you folks have been checking out this thread in Pirate, I may need some help if that one guy posts a question over here.

You know the one.

The guy that uses the law of conservation of energy to explain why there is less chance of an arc on the ground side since all the electricity has been used up by the load and therefore there is zero volts on the ground side. And therefore no arc.

I know the answer, of course, but I was awake half the night last night trying to come up with a way to explain it so he can understand.

Maybe if I explain to him that the electrons flow from negative to positive......Nope, that will open a whole nother can of worms.

Plus, there is another guy, or maybe its the same one, that seems to think that ground is the place all good electrons go to die.

Another guy says that ground switching is better, and that all cars used to use it to prevent arcs.
You think that explaining the difference in cost between one wire and two would help?

Gotta go, there is a wire EDM machine that is giving a constant ground signal that a tool and die shop need fixed.

J.
 
Nice story Fordum.. But you didn't fix the OP problem.

How about just putting in a normal DC bulb and socket for the indicator light!!!
 
Oh, that would probably do it, there probably is not much current involved. But he was wanting to know why it was lighting, and where the voltage is coming from.

Given the circuit shown, there are a couple of possibilities. That is why I suggested that if he wanted to do some checking, we could probably figure it out.

Heck, all he has to do is measure the voltage across the LED at maximum brightness.
(When it is not supposed to be on) and put a Zener Diode of about that voltage in series with it.
That would not let it see any voltage until it got above the level of the Zener.

J.
 
Lunatic said:
So before I even start, would anybody care to shed some light on the subject of actually connecting the electrical circuits for this fan ?
Click to expand...

I'm covered, excused - And proud of how far I've come from nothing-ness. :haha: I went from zero to functional - Granted, in a non-educated manner, which will (and has) presented flaws in the overall design.

With that being said, Fordum, thanks for the compliment of my fabricating madness.

Now, to your actual response.

I dont mind those which show a heavy hand (in your words). It shows character and how much a person believes in what they are suggesting or commenting about. Actually, I should thank you ! You seemingly have started a fire here (Which is what I believe your comment of regret was about). And with this fire, I'll actually get a solution or at least more ideas of a solution. :bow:

The distribution blocks are (yes) steel. Perhaps not the greatest conductor. The positive block is on a list of things to replace (I dropped it, if you didnt catch this in an earlier post). I did repair it back to original condition - But it just isnt flawlessly solid (in my mind) anymore. Now I have the reasoning to replace both blocks, exact same design of course, but using copper as the conductor. The steel is actually encased - So, adding a strip across the top - Nah, it would definitely look as it was - An after thought.

It looks like you have the anode of the indicator diode on the common, along with terminal 85

If Im understanding what you're saying - You are correct, and the diagram is (discovered a week or so ago) wrong. The power for the blue LEDs is actually coming from (actual, physical connection) a fuse block tap terminal on the ignition circuit, fuse block. It so happens, however, that the switches are getting power from the exact same place.

SPST lighted switches.

When the switch is on, but the coolant switch is off, the LED is seeing 12 volts on both sides of it. Ignition switched 12v on one side, and 12 volt through the fan from the terminal block on the other.

I believe you are correct on this as well. Im not questioning what you have stated - I think I remember measuring voltage in this very loop the very first time I wired everything together.

Or the fan rotating is generating voltage and back feeding the strip

I think this is the actual problem. Although, actually thinking about it confuses me. If the fan motor is backfeeding the grounding circuit with power when it "wind spins", wouldn't that power generated meet the iginition power on the other side of the blue LEDs - Leaving the blue LEDs un-lit ?? Perhaps you can explain this.

Im fluke in hand - What do you suggest checking first ? :D


owenst7, BGKK5 - Kirchhoff's Voltage and Current Law....... ? There goes simplicity.... :dunno:

90k5blaze, Wouldnt a light bulb function just the same though ?

....and finally..... Fordum, I'll check the voltage and post it here.

....all he has to do is measure the voltage across the LED at maximum brightness - When it is not supposed to be on - And put a Zener Diode of about that voltage in series with it. That would not let it see any voltage until it got above the level of the Zener.



Thanks for all the help and insight. Hopefully a solution presents itself by the end of the weekend.



Troy B
Ft Hood, TX
 
OK, I am going to try to explain my idea. Its not a question of "dumbing it down" for you, your're not dumb...its a question of not using technical terms or concepts that you are not familiar with.

Lets try this:
All conductors have resistance, except for superconductors. Because of this resistance, there will always be a voltage drop from one end of a conductor to the other under load.
As a result, the voltages in the wires in your truck will always be ever so slightly different from one wire to another.

Normally this is unnoticeable.

But, if you have a poor conductor, the voltage drop is increased.
Its possible, that the steel strips you used might be causing this problem. I don't really think so, but its a thought.

A quick way to test, would be to disconnect the positive and negative wires from your fans, and connect them to the same posts you are getting the positive and negative from your battery.
Thus bypassing the strips.

If the LEDs quit glowing when you drive, you have found it.

Before you do that, try this if you can.

Short the temp switches so that the fans come on without the engine being hot or even running.
Put your Fluke on the low VDC reading, probably 2 volts if its not auto-ranging.

Measure from the farthest positive terminal on the strip to the positive battery wire connector.
If you see a voltage, that is half the voltage you are losing due to the resistance of the strip.
The negative strip is losing that amount too.

Since the strip is supposed to be a dead short, you should see zero volts.

Or, put it on 20 volts, and measure from positive to negative fan connections on the far end of the strip.
Note the voltage and measure from the positive to negative on the battery wire terminals.
Should be the same. If not, that is the voltage drop of the strips.

All these measurements must be done under max current load. (both fans running.)

Now, all that was to determine if you have a voltage drop problem with the strips.
I wrote all that down, to stall while I tried to figure out how to write this next part.

I think I have figured out how to write it.

The fans are acting as generators. They are producing voltage in the SAME direction as they are normally supplied.
Here is the part that might give you trouble.

Since they are not hooked to ground, with the relay off, they are INDEPENDENT of the battery and alternator.
They share no ground reference with the rest of the truck.

Now, trace that circuit with your mind.
I am not going to get into the voltage direction of travel thing.

The power goes from the fan/generator, out the positive wire to the terminal strip. From the strip to the positive post of the battery. It sees no battery voltage, because the negative terminal is not in the circuit anywhere.
It goes through the positive terminal to the ignition circuit. From the ignition circuit to the anode of the LED, through it to the negative wire of the fan/generator completing the circuit and lighting the LED.

If it helps, think of the fan as a battery with positive and negative terminals. When you hook the LED across the battery, it lights.

If you have a way to spin the fans, I suggest an air hose, just don't put too much unbalanced load on the bearings and don't spin them too fast.

Disconnect the LED, put your meter leads from the positive wire to the negative wire of a fan and spin her up.
Under no load, you might hit 12 volts.

Gotta go, its been another long day, and I'm crashing hope I made some sense.

J.
 
WOW!!! Fordum's post 62 and down is why I will never understand electrical. WAAAAAAY over my head. I need to show this thread to a buddy of mine who does AMAZING electrical work and LOVES electrical work. He would absolutely love reading this kind of stuff. My eyes roll in the back of my head reading most of this. I just don't get. I'm just NOT smart enough to. Maybe because I've not been schooled on electrical, I guess. Give me a block of steel, aluminum, or any kind of alloy and I can machine it, throw grinders at it or flame at it and turn in into something really cool, but wires....wires frighten me. You guys are amazing. :bow::bow::bow:
 
You mention replacing the strips because adding copper would look afterthought.
My idea was to make a thick strip of copper with the same holes as the strips, but without the threads. Slightly larger so the bolts will pass through.
Make the dimensions slightly smaller in length and width so the color underneath would show. Then get some jeweler's rouge and polish those strips until they were like a mirror.
Remove all the wires. Using cotton gloves install the strips over the bolts, reinstall the wires, tighten them down and put several coats of clear Krylon or some other kind of sealer to stop tarnish.

Or, even better, get a kit and gold plate them. I know gold is high now, but a few microns would hold a shine forever and not be too expensive.

J.
 
why not just buy a bus bar?

I can certainly appreciate the effort and patience to build the distribution blocks, but all it really is is a homemade bus bar... nice marine ones can be had for $10 to $100 depending on amp rating...
 
Mainly because he wants to, I imagine. Same reason folks build their own bumpers, dashes, and what have you.

Sometimes because exactly what you want is not available, but most times just to say " I did that".

Of course, lots of times you just have to hang your head, scuff your feet, and say "Yeah, I did that "

J.
 
well, I'm just not sure if he knew, what he wanted, was available for purchase.... you can get just about any configuration...
 
Fordum said:
The power goes from the fan/generator, out the positive wire to the terminal strip. From the strip to the positive post of the battery. It sees no battery voltage, because the negative terminal is not in the circuit anywhere.

It goes through the positive terminal to the ignition circuit. From the ignition circuit to the anode of the LED, through it to the negative wire of the fan/generator completing the circuit and lighting the LED.
Click to expand...
　
......And there it is......I also made sense of that. The question now is - Is there a way to break the circuit (immediately) once the relay breaks the contact ?
　
Fordum said:
My idea was to make a thick strip of copper with the same holes as the strips, but without the threads. Slightly larger so the bolts will pass through.

Make the dimensions slightly smaller in length and width so the color underneath would show. Then get some jeweler's rouge and polish those strips until they were like a mirror.

Remove all the wires. Using cotton gloves install the strips over the bolts, reinstall the wires, tighten them down and put several coats of clear Krylon or some other kind of sealer to stop tarnish.
Click to expand...
　
Thats a good idea, actually - I was stuck with a very basic idea. Most likely because I am so set on re-making and re-designing the blocks.

I think I'll use the external (visible) bar idea with the next set I make - It will make the fabrication of the actual blocks much simpler - And more than likely look better than what I have now.
　

Ryoken, your question of why ? (Not being an a$$hat here or anything), you could have bought how many items for your build, but chose to fab them yourself ? Your console, for example - Wouldnt be that particular console had you just dropped a dime on one that would suffice even though theres so many out there.

My blocks are of the same idea - Just on a much, much smaller scale. Enjoyment of the fab work, pride in the finish, and a built myself coversational piece - no matter the key component in question. True ?

Thanks for mentioning the availability to ensure I knew though. :waytogo:



Thanks everyone for the continued digestion of this topic - Comments and further questions / suggestions always welcome !



Troy B
Ft Hood, TX
 
Last edited:
Just clicked on this thread for the first time today..... the subject line does not do it justice.

Initially I thought "oh boy...ANOTHER Windstar fan thread :rolleyes:" but the contents of this thread transcend it's humble title.

T-RexAwesome.jpg



Subscribed for the final photos and the insane propeller-head technical discourse!


:saweet:




:usaflag:
 
I have to agree with the incredible anal retentive attention to detail.:bow:
I know Greg and identify...:D

VERY nice work!

Only critique is the use of the crimp-on terminals.
A nice soldered and crimped connection with some quality heat shrink would finish it up nicely.

Nice fab work.
 
American84k5, thanks for dropping in and posting the kind words :waytogo:

Greg 72, I'll address that issue (title) later today I guess - Something "catchier" (if thats even a word).

BobK, Im not too crazy for crimping either. Perhaps when its all said and done I'll pull out the solder - A final, final step.


I have something else up my sleeve that only one person knows about and has actually seen in person. I may post a pic of it later as well.

I'd really like to nip this glowing issue in the bud first though. I still lack glassing a shroud and finish coating the brackets before this little project is completed.

I thought I'd be pulling my water pump and timing chain today, to finally replace them with a noisey direct drive and electric water pump. The glow issue is making me stop though. I dont want to add another circuit to the relay panel, only to have to move more wires aside and what not (should it become necessary). Instead, I guess I'll just be re-seating my steel trans oil pan.....again.... :( ...... being that Summit's aluminum pan is out of stock until the 6th of June.

I dont want to move forward until this is resolved, it's that simple - Im tunneling on this one issue atm.





Troy B
Ft Hood, TX
 
there's absolutely nothing wrong with crimped connections, when a decent terminal is used, and it's done correctly... soldering has it's downsides too...

the key with crimping is to NOT use the cheesy hard plastic ones.. either buy a quality one with good adhesive lined shrink on it, or raw terminals with your own quality shrink tube... the proper style crimp tool for that terminal is critical too...


and I understand fully about the bus bars... just wanted you to be aware that what you want is more than likely available.. even I go for a stepsaver here and there..... :doah::wink1::haha:

impressive attention to detail, me likey! :bow:
 
You must log in or register to reply here.

Latest Posts

Top Bottom