Tony

you have that whole shroud/assembly...or just the fans?

Lorenfb

A total waste of time and money! Bosch did an excellent and very conservative design. Seems as if many don't understand how the fan controller and amplifier function together,
e.g. how the controller drives the amplifier.

jej3

I used the whole shroud assembly. From memory, I cut mounting tabs off, notched the bottom of housing to attach to radiator and used straps to attach to upper radiator mounts. It is very straightforward and is less deep than the original fans.

Many folks pulled these same fans off a Ford Contour, I think. I just preferred new, at that price. No pics at the moment.

https://rennlist.com/forums/928-foru...ht=dorman+fans

Tony

I really don't care how they work together as long as they work like they are supposed to.

Until you source replacements..for $300-400 amps..controllers...then fan motors. Each his own.

Lorenfb

First, the controller (silver box) basically never fails. The power module (black with fins) has a failure rate less than 1-2%.
As mentioned before, most don't know how to troubleshoot the cooling system and usually assume it's the power module.
The power module is extremely simple to test, i.e. a ten minute effort worst case. Used/tested modules are available
for less than $150.

jej3

Curious how you know what the failure rate is?

Lorenfb

Tested many and only needed to repair one; http://www.systemsc.com/products.htm

Alan

Agreed the controller box seems very reliable. Final stage units much less so (rather understandably). You say 1-2% failure rate - but failure rates are usually expressed over time - hence MTBF etc. Are you suggesting only 1-2% in the lifetime (so far) of all cars so equipped? or 1-2% per year - or something else?

My own observations of Rennlist traffic on this and personal experience suggest it must be much higher than 1-2% lifetime (so far). I do still think the modules are acceptably reliable given their duty and easily enough repaired that a replacement design is not a high imperative. If a fan stops working it is most likely the fan itself or connectors/wiring at fault.

Alan

hans14914

Its unbelievable how robust the factory system actually is. Especially when you consider the design of the output stage, and how much current and heat those gates have put up with. I got my scope out the other day to start logging the control signal. I think that maybe just an updated circuit board and a bit simplified layout would be pretty easy project. I already reverse engineered the connector, so I can make those.

I gather its a fairly simplistic duty-cycle style control, but have no idea what the frequency is yet, or how it relays back to the control that "I have a bad fan, run the other on full". I see the wire that does this, but need to watch that behavior to see what its sending back.

Lorenfb

So you have personally tested the all "failed" power modules noted on Rennlist, right? Or have you forgotten the most don't know how to troubleshoot the 928 cooling system,
and the power module is the easiest to assume as the failure point?

Alan

You have a fine way of picking arguments with people who are basically agreeing with you...

Lorenfb

So why the need to re-engineer any part of the 928 cooling system? Maybe some need to view the schematics and read the Porsche manuals, to understand its functionality.

Correct.

It doesn't. The is no processor in the power module, it's "dumb".

hans14914

Loren,

That is interesting. From reading the manual, and experience, there are failsafe modes that call the fans to run at 100% and also check the fans for non-running conditions. I interpreted that it should check the fan (assuming a current draw) and if it was out of spec, it caused the other fan to run at 100% and waited a period before testing the offending fan again, and then if rectified, it would return to its temperature/conditioned referenced speed.

I made the assumption that this logic was being done inside the controller, as I did not see any comparators or logic devices inside the output stage. However, I suppose it could be done in there with some simple gates. I have not had the time to sit down and draw a schematic from the PCB yet, and have not found one published. If you happen to have one, I would love to take a look.

Lorenfb

There're only three outputs from the controller to the amplifier; UPow (12V power) pin 7, Sig1 (PWM) pin 6, Sig2 (PWM) pin8. Pins 6/8, voltage doubler (24-26 volts) outputs, to provide
gate drive voltages for the MOSFETs to saturate them - reducing their power dissipation.

There're some ICs in there to do some minor controls, e.g. monitor temp & max current.

jleidel

Some people should have their internet access revoked.