|
Re: Mercedes fan upgrade
[Re: Andrewh]
#838392
03/05/11 07:43 PM
03/05/11 07:43 PM
|
Joined: Oct 2007
Posts: 3,896
Oregon
hooziewhatsit
master
|
master
Joined: Oct 2007
Posts: 3,896
Oregon
|
I'll have to see what happens if I drive the original trigger wire with 90%, and the other wire with the varying duty cycle.
On my current test board, I have a tiny relay that ties the original wire to 12v while the other wire controls the speed. To turn the fan off, I switch both inputs to ground.
If you ever find yourself in a fair fight, your tactics suck.
|
|
|
Re: Mercedes fan upgrade
[Re: hooziewhatsit]
#838393
03/06/11 06:51 PM
03/06/11 06:51 PM
|
Joined: Oct 2007
Posts: 3,896
Oregon
hooziewhatsit
master
|
master
Joined: Oct 2007
Posts: 3,896
Oregon
|
I've been talking with Andy through PMs about a way to use the stock sender, and have the controller move the needle on the temp gauge in the dash. What I've found is... With a solid state (constant) 5v voltage regulator on the back of the cluster, I can move the needle though its range of motion. With the stock pulsing regulator, I can only get the needle to about 1/3rd on the gauge. So, if you have a solid state regulator, you can hook the stock sensor into the controller board, which will use that to control the speed of the fan. Then, you hook the wire that used to go to the sensor, into the controller board, and the controller will move the needle to show the temperature. It may just be easier to drill and tap a third hole in the intake 
If you ever find yourself in a fair fight, your tactics suck.
|
|
|
Re: Mercedes fan upgrade
[Re: feets]
#838395
03/07/11 12:16 AM
03/07/11 12:16 AM
|
Joined: Feb 2003
Posts: 8,684
Andrewh
master
|
master
Joined: Feb 2003
Posts: 8,684
|
wow, I see what you mean about drilling a new hole. I can try to build one for about 10 bucks. diy regulator buy an odd looking one for 30 bucks. better than diy or spend 50 on a high tech direct replacement. direct replacement or just use a drill and tap I have to put in another. OR I guess I could live without a gauge in the car. Nah, will proably try to build my own. so how soon for a prototype to try? and did you see if you can drive both at 90 percent, the trigger and controler wires?
|
|
|
Re: Mercedes fan upgrade
[Re: Andrewh]
#838396
03/07/11 04:54 AM
03/07/11 04:54 AM
|
Joined: Oct 2007
Posts: 3,896
Oregon
hooziewhatsit
master
|
master
Joined: Oct 2007
Posts: 3,896
Oregon
|
I have not seen if I can drive both wires at 90% and have it still run. I guess I'm not too worried about it. If all of a sudden it's on full speed all the time, there must be a break in the speed wire, or the controller is fried. Either way it warrants a look to see what's going on. The circuit will be able to drive it at 90%, or 100%, so I can figure it out later, as my current prototype can't do it. ETA for another prototype... hmm... I guess it depends on how much more we're going to change the design  I know the circuit I need to control the speed of the fan and to read the temp sensor. I also have a circuit that will move the needle on the gauge in the dash. I think we have everything pretty well nailed down, so I could go ahead and order some circuit boards. The alternative is $100 for a development board, and build the entire circuit to make sure it all works. Oh yea, Andrew, what are your thoughts on the dip switches I used on the boards I sent you? Too hard to switch/too annoying/easy enough?
If you ever find yourself in a fair fight, your tactics suck.
|
|
|
Re: Mercedes fan upgrade
[Re: Andrewh]
#838398
03/07/11 03:45 PM
03/07/11 03:45 PM
|
Joined: Oct 2007
Posts: 3,896
Oregon
hooziewhatsit
master
|
master
Joined: Oct 2007
Posts: 3,896
Oregon
|
As my wife succinctly put last night, I can do the dev board for $100, fix any issues, then get circuit boards for another $100 ($200 total). Or, get circuit boards for $100, and either be done, or fix issues and get another set for another $100. So likely only $100, and maybe $200.
I'll start the new layout today and send it off when I can. It'll be another ~2 weeks for a prototype. I also agree that we seem to have all the features nailed down. At this point it's just software to logically control the speed.
The new boards will probably have another dip switch or two on them to indicate GM or stock mopar sensor and whatever else I come up. Will probably include a switch to disable variable speed?
I'll probably make the other boards available at a pretty good discount for those that don't want variable speed.
If you ever find yourself in a fair fight, your tactics suck.
|
|
|
Re: Mercedes fan upgrade
[Re: Andrewh]
#838400
03/17/11 05:39 PM
03/17/11 05:39 PM
|
Joined: Oct 2007
Posts: 3,896
Oregon
hooziewhatsit
master
|
master
Joined: Oct 2007
Posts: 3,896
Oregon
|
Alrighty, any other features I need to add? I'm almost done with the new version of the circuit board. I was hoping to be done with it a week ago but some stuff came up.
This version adds variable speed and the hardware to move the needle on the stock temp sensor. It will also bolt into the bigger white box using two standoffs in it (with two more sticky standoffs to support it).
I decided to remove the output to the dash status light. I really don't think I'll use it, nor would anyone else. I can hack one in later if someone really wants one.
If you ever find yourself in a fair fight, your tactics suck.
|
|
|
Re: Mercedes fan upgrade
[Re: Andrewh]
#838402
03/18/11 02:42 PM
03/18/11 02:42 PM
|
Joined: Oct 2007
Posts: 3,896
Oregon
hooziewhatsit
master
|
master
Joined: Oct 2007
Posts: 3,896
Oregon
|
Quote:
so, now you can
control the speed.
use either stock temp sensor or your aftermarket.
run the fan on or off, instead of speed control.
feed the stock gauge with a constant 5v vr for the gauges.
have a check for loss of temp sensor(either one?)
still need to confirm 90 percent driving either on/off or loss of temp sensor.
a/c signal
adjustible on by temp
anything I forgot?
I think you got everything 
To clarify for others, "feed the stock gauge with a constant 5v vr for the gauges.", if you have a constant 5v reg in the cluster, I can vary the ground of the gauge to move the needle. The controller board itself does not provide the 5v.
I can detect a loss of either sensor and turn the fan on. I still have the dip switch to allow a relay or switch to turn the fan on.
Surprisingly, the added hardware for variable speed and to move the needle only added about $5 to the total cost.
I may be able to finish the design and send it off today, but with family coming tomorrow, I may need to wait until monday. 
If you ever find yourself in a fair fight, your tactics suck.
|
|
|
Re: Mercedes fan upgrade
[Re: Andrewh]
#838404
04/02/11 06:51 PM
04/02/11 06:51 PM
|
Joined: Jan 2003
Posts: 1,476
gtx69
super gas
|
super gas
Joined: Jan 2003
Posts: 1,476
|
Do you have the variable fan controller ready to go?I'll be ready to ride in the next month.
Last edited by gtx69; 04/02/11 06:58 PM.
|
|
|
Re: Mercedes fan upgrade
[Re: hooziewhatsit]
#838406
04/03/11 05:56 PM
04/03/11 05:56 PM
|
Joined: Jan 2003
Posts: 28,068
Irving, TX
feets
Senior Management
|
Senior Management
Joined: Jan 2003
Posts: 28,068
Irving, TX
|
Good news! Now I can start working to convince you that updating the Auto Temp systems is in your best interest. 
We are brothers and sisters doing time on the planet for better or worse. I'll take the better, if you don't mind.
- Stu Harmon
|
|
|
Re: Mercedes fan upgrade
[Re: feets]
#838407
04/10/11 02:13 PM
04/10/11 02:13 PM
|
Joined: Oct 2007
Posts: 3,896
Oregon
hooziewhatsit
master
|
master
Joined: Oct 2007
Posts: 3,896
Oregon
|
Quote:
Good news!
Now I can start working to convince you that updating the Auto Temp systems is in your best interest.
Show me the market and all the specs, and I'll put it in line behind the fan controller, 46RE controller, and Andy's seat controller.
Got to the point I could test the fan this morning, and it seems to work just fine 
With a 90% signal on the trigger line, and a signal on the speed line, it doesn't work as desired.  With a speed signal, it appears to require a 100% signal on the trigger line. So if the fan suddenly goes to 100%, check the wiring to make sure everything is still connected.
I still need to test moving the needle with the new hardware, but otherwise, variable speed control works Now I just need to come up with the control loops to vary the speed according to temperature.
If you ever find yourself in a fair fight, your tactics suck.
|
|
|
Re: Mercedes fan upgrade
[Re: hooziewhatsit]
#838408
04/15/11 10:14 AM
04/15/11 10:14 AM
|
Joined: Jan 2003
Posts: 1,476
gtx69
super gas
|
super gas
Joined: Jan 2003
Posts: 1,476
|
Quote:
Quote:
Do you have the variable fan controller ready to go?I'll be ready to ride in the next month.
The circuit boards should show up on Tuesday (4/5). Another week or so to get the prototype going and another week or two for testing... I'd SWAG they should be ready in about a month
  
|
|
|
Re: Mercedes fan upgrade
[Re: gtx69]
#838409
04/20/11 09:58 PM
04/20/11 09:58 PM
|
Joined: Oct 2007
Posts: 3,896
Oregon
hooziewhatsit
master
|
master
Joined: Oct 2007
Posts: 3,896
Oregon
|
Welll... had another urgent project come up that took a good chunk of my time the last couple weeks. Good news is, if that project takes off, I can retire early and spend more time on fun stuff like this Anywho, I have the new code more or less tested in Excel to vary the speed according to the current temperature. Now I just need it to stop raining so I can install the fan, and/or, the seller would stop jerking us around so we can move to our new place that has a garage Oh yea, the prototype hardware is mostly done, and can control the speed just fine 
If you ever find yourself in a fair fight, your tactics suck.
|
|
|
|
|
