Alan

Ahh - good idea - the wiper parking circuit - hadn't thought of that one!

Yes and this is actually a good solution for all years, exists on every model.

Getting from the cowl area is easy - just use it as a relay trigger though (as here) - not as a supply. I'd still fuse this trigger wire at the motor end with a 500mA or 1A fuse so there is no risk of blowing the wiper fuse due to any issues on this... that would be bad.

Alan

jwillman

Thanks for the input Alan. I will report back.

dr bob

Jim--

You can pull the relay and put in a fused (small fuse, like 1 or 2 amp) jumper between the two relay coil contacts in the CE panel socket. That will give you battery+ at the blue-green wire when the ignition is on. The diagram shows the blue-green connected to both the engine coolant temp switch and the AC temp switch. You can leave those connected, and remove the brown wires from those switches. The now-open side of the switches can now be used to trigger your new fan relays. You'll have the option of playing diode-logic games with those relays to get combinations working. For instance, you may want to use 1 fan only on coolant return temp greater than initial warm-up temp. Both fans when engine gets warmer, .or. when AC is on.

-----

I received an interesting e-mail from a listmember who's working to get his late S4 fans to come on sooner, so his new lower-temp thermostat will try to keep the gauge lower. It reminded me that the temp sender location in the cold-side header means that the sender or switch installed there only regulates the temp of coolant going back to the engine from the radiator. Lower return temps obviously have a possible effect on gauge temps, but it isn't a direct connection by any means. Returning coolant flows through the pump and most of the block before it reaches the temp gauge sender in the water bridge, so the gauge really indicates the result of the returning coolant temp plus the added heat plus the effects of the bypassing thermostat to the bridge temp. Bottom line is that adding fans earlier won't necessarily lower the gauge reading; it just causes the engine to warm up more slowly.

jwillman

Thanks Dr Bob.

The Hayden Dual fan controller I got (Orielly's) has two 40 AMP fan relays each with its own fused power supply wire which I connected to the jump post. The relays are designed to run up to two fans with two speeds. In my case I am using the seperate relays to run the two single speed fans independantly. The assembly has a temp prob which inserts into the radiator fins up near the hot water intake and fan 1 turn-on is adjustable via a temp pot on the relay assembly. The second relay is factory set to turn on fan 2 at 10 deg higher than the fan 1 setting. I liked this idea as it seemed to help with voltage drop / surge concerns versus both fans kicking on at the same time.

I think I will leave the AC fan set up as is and get my ignition voltage signal from the wire coming of the X circuit going to the wiper.

dr bob

Jim--

You will probably want full fan operation on AC run.

Jury is out on sampling coolant themp at the hot side of the radiator. I know lots of folks do it, but from a process and parasitic load point of view it's hardly the best choice. In my opinion anyway.

jwillman

The fan controller also has a signal wire that runs to the AC clutch to run fans when ever AC is on. For now I was not connecting it because I wanted to see how the temps responded to new set up versus the dying viscous fan.

I did a test with the ignition signal wire connected to the hot post this weekend and temps never got to the second white line at idle. As soon as the first fan kicked in they started to drop immediately. All my viscous fan could do was keep it at the 2nd white line.

Where would you recommend the temp probe go if not by the hot side intake? Should I compensate for those concerns by adjusting when the fans turn on?

j.kenzie@sbcglobal.net

Isn't one of the coil wires a switched 12 volt source? The fan controller doesn't use this for power, just to know condition of ignition power.
Dave

svpmx83

Another good point to tap into on the CIS models is the positive feed to the WUR or Suplementary Air valve.
1. its available from the 14 pin connector
2. it's the same feed as for the fuel pump so is only 12v when the engine is running.

jwillman

Thanks guys.

So is circuit 15 also ignition switched then?

I am a bit concerned about tapping into my ignition / fuel control circuitry but if circuit 15 is also ignition switched those are options. The 12V supply wire to the transitor ignition unit (square metal box below jump post?) is inches away from where my fan controller is mounted, short wire run!

Alan

15 is an ignition switched supply in the purest sense - it is on whenever the ignition is on - including during starter cranking...

X-Bus switched is really better for the fans - they can take a lot of current and you don't want them to also be on during starting... e.g. a hot restart.

Alan

dr bob

Most folks seem to add fans to keep the gauge where they want it, yet the fan systems in the market seem to want to control fans based on the temp of the water entering the radiator. In a perfect world, the fan controller would have a map that predicts heat load and heat shedding capacity. Kind of like the way the LH works for fuel management, but substitute a temp reading from the water bridge for the oxy sensor input. That's not happening though. Turns out I spend quite a bit of time doing this sort of control in various industrial systems so maybe I'm overly picky.

Install the kit per the seller's instructions and enjoy it. The thermocouple they supplied sees the amount of heat radiated and adjusts the fan operation when that gets high. Ultimately the fans run way too much, but it's probably OK then letting the water thermostat regulate engine temp and, as a fringe benefit, keeping the gauge somewhat consistent.


More current 'murican car cooling systems use the water bridge temp to control the fans, a shared sensor doing the measurement work. The control is more than a simple feedback loop, in that they integrate the fan conrols with the whole engine management package. This is much more predictive of heat load changes, the first half of a full predictive model. Next time we are designing a new engine management system for our vintage cars, we should be sure to include this feature.

I think that Hacker once mentioned an all-fans temp control system that Z or Tim Murphy is using. Might be interesting to get some details o that system and how well it's working.

tveltman

Arduino has a built-in PID library if anyone is interested in tweaking it to make it work. You know, just F Your I.

jwillman

Thanks all for the input.

Used the bk/gr wire to the wiper for 12V switched and my electric cooling fans work like a charm. Marked inprovement in operating temp at the gauge over my worn out viscous fan.

BC

Tell me more about that! Load per TPS and AIT and duty cycle determines what will happen with the fans?

Its working very well - and while wisconsin has summers that rarely see 100F, it gets very humid and in the 90s. Add in the load of a centrifugal SC running up to 30psi as I recall, I think its a great test of the theory.

The applied science and hardware is no thermostat. Yes - in the 928 you cannot simply remove it. You must block off the back hole on the bridge and allow the circulation through the front area. This was done with a disemboweled stock T-stat.

The fans are not stock, and it is recorded in lore that Todd's car could probably "suck" its way forward in neutral with both fans on full. It needs this because of the a/a intercooler in front of the radiator.

Caveat - I believe he runs two alternators.

dr bob

I'd be tempted to just look at injector pulse width to get a handle on the load side of the prediction. Lag the 'decrease' side of the speed profile a bit and it would be darn close. Trim with measured coolant temp. Not really rocket science, especially if I can do it.

On the all-fans solution, it would take a while to warm up with no thermostst recirculation while cold. Need to maintain some flow so there aren't any hot-spots; Some sort of full-time thermostat bypass is the conventional way, while Porsche decided to meter the recirc flow for warm-up. Combination of the two methods works OK, but it runs the fans more than they really need.