• Welcome to The Truck Stop! We see you haven't REGISTERED yet.

    Your truck knowledge is missing!
    • Registration is FREE , all we need is your birthday and email. (We don't share ANY data with ANYONE)
    • We have tons of knowledge here for your diesel truck!
    • Post your own topics and reply to existing threads to help others out!
    • NO ADS! The site is fully functional and ad free!
    CLICK HERE TO REGISTER!

    Problems registering? Click here to contact us!

    Already registered, but need a PASSWORD RESET? CLICK HERE TO RESET YOUR PASSWORD!

Designing A 3 ECT Sensor Setup for the 6.5

J

JayTheCPA

Well-Known Member
Messages
2,350
Reaction score
2,882
Location
Annapolis, MD
Have three ECT sensors (thermostat housing, passenger head above #8, and driver head above #1) as I prepare for installation of a ECT controlled fan.

OE configuration:
> Thermostat's sensor sends to the ECM.
> Driver's head above #1 sends to the dash gauge.


With three sensors, am considering this setup:
> Thermostat's sensor sends to the fan.
> Driver's head above #1 sends to the dash gauge.
> Passenger's head above #8 sends to the ECM.

Any downsides to this?
 
Put the fan switch over #8. Typically you want the cooling fan switch in one of the hottest spots, and #8 is a known a hot spot.
 
For the sake of discussion (read: am not disagreeing), isn't it a little late by the time #8 gets hot? IIRC, the thermostat housing reads coolant coming out of the radiator, and presuming so, does it make sense to catch the flow as early as possible rather than mid-way through the motor?

Also, should have mentioned that I have a restriction in the bypass hose.
 
#8 is a hystorically known hot spot in GM's V8 engine design. Small blocks, big blocks, and even the 6.5L have all sufferred from this. This was why they came out with the balanced flow water pump to help with it. I know I've seen a 30 degree difference from the drivers side front port to the passenger side rear port on small blocks with #8 being 30 degrees hotter. It also showed higher than the t-stat housing does. The housing is the last stop before the coolant goes to the radiator, but is not always the hottest spot. #8 also suffers from the fact it is a spot where coolant tends to sit idle. It's in the back of the head, so most of the flow is happening in front of it. The t stat housing will be the most stable temp reading, hence why GM put the CTS there. The ECM needs to see a steady temp reading that is fairly representative of the engine average temp, and the flow at the t-stat provides this. Try em out, but I would not move the CTS from it's location as the fluctuations could be bad for fueling AND shifting since the ECM adjusts shifts according to engine temps in certain temp ranges, but I would almost be willing to wager that the back passenger head runs hotter.
 
Ah, so from the thermostat housing the flow goes to the radiator? If so, then agree that this is NOT the place for the fan's sensor.

What are the thoughts on adding a sleeve to the return-back-to-the-motor hose coming from the radiator and controlling the fan that point?


sctrailrider said:
Ment to ask what fan clutch you planing to use..
Click to expand...

Not 100% sure yet.

The electro-magnetic and hydraulic ones need further thought as I do not want to run them at 100% and load the motor just cause the A/C is On. To avoid loading while the A/C is On, these will need pusher fans out front and modification to the grill. Advantages are that they can freewheel (minimize parasitic drag) and let the motor warm up faster.

The electro-viscous fan just needs a few test fits to see which diesel rated fan mates to it and whether the fan fits in the 6.5's shroud. Advantage is that the speed is controllable based on ECT and likely will not need a pusher for A/C use.
 
I was thinking the bottom radiator hose would be best place for triggering fan but after writing a bunch of stuff I changed my mind. So lets discuss it some more and see which control you want to accomplish as that might help the decision.

Yeah, the thermostat measures an average temperature of both heads and some ? circulation up through the head gasket from around the cylinder's water jacket. The water recirculates in the engine some too. Water flow in any open or multi path system will take the path of least resistance. It is sometimes funky and tricky so some areas are hotter than others. So that is not a good trigger point for upper limit panic point.

You know this just rambling..... The thermostat is suppose to maintain the minimum engine temperature but it can't maintain upper limit. It opens open over small window of few degrees and opens to max at a few degrees above that. That is the limit of its control other than it will modulate if it has to maintain minimum. The fan control is trying to help modulate the temperature too.

The difference in upper to lower radiator water temperature and depending on flow rate has to match the heat load of the engine (the coolant portion of heat load). I don't trust my memory but maybe it had to be a minimum of 20 degrees difference for the radiators I was testing. Such that if the radiator size and fan could not cool top to bottom hose difference more than 20 degrees the engine temperature would start to climb and run away. We would have to change air flow, radiator size, and or once contemplated water pump speed/flow (thought of pulley change but changed something in the stack).

So I was thinking if the lower radiator hose goes above ~ 160-170 it will run away past danger zone? In which case the fan should come on max but what warms the lower radiator hose its the engine and its farther downstream from the heat source and big average. It will also take a fair amount of time to react and reach danger zone so you might get behind the Q ball trying to save the engine.

They try to engineer the radiator not to overcool the engine. It saves money on radiator size and you don't want a cold water jacket and hot heads and the thermostat constantly cycling.

A lot of rambling and a bit more if you want to optimize it. It might be good to trigger at different temps at different locations for different reasons. And see how it reacts.

It also depends on ambient and how "strong the heat load is coming on".

What were you thinking? Depending on your thermostat temperature. Assuming its 195F.

1. I am thinking come on 50% fan speed if average thermostat temp reaches 200F or at rear head 220F.

2. Then max at 210F average thermostat temp or rear head 230F which ever first ????

Sorry this is getting exceeding long. I am trying to say I think you need 2 types of controls a normal and safety.

For hot days light load key off avg thermostat temp say at 200F half speed fan.

Safety for what ever reason anytime if rear head reaches 230F ??? Turn on fan to max.

What say ya'll.
 
Light steady load high ambient temperature condition the bottom radiator hose or a stack temperature might be a good parallel point to trigger a soft fan speed . This might help get ahead of heat onset and not require full fan speed depending on trigger temperature???? But again not a good place for triggering an engine temperature safety if you have other options.
 
Excellent ramblings :)

Agree that without the benefit of a ECM / system that accepts multiple inputs, we are left with picking a single point of measurement and base the logic (or in this case, the mechanical reaction(s)) from that point.

How is this thought: Control from the hottest coolant going to the radiator while still getting dash / gauge measurements from at least one other point (ex: above #8). Reasoning:
> this is the volume of coolant that the fan is able to directly affect.
> if the fan cools the fluid too much, the thermostat(s) will maintain minimum temperature within the block.
> when things are getting hot, the fan is quicker to react and starts bringing down temperatures *before* the hot coolant even enters the radiator.
> if the fan is 'locked-in' while pulling hard and the dash gauge is showing that things are just too hot, this allows the driver back-off of the throttle when / if there is simply more heat than what the system can shed. Ditto if the thermostat(s) fail closed.

Got-it that above the #8 is probably going to get the hottest under load. At the same time am just not sold that a fan trigger at this point is optimal as the fluid temp is going to rise / fall relatively in sync with the rest of the motor (read: it is a little late to react from this temperature). Am convinced that a dash gauge that shows temperature above #8 is worthwhile as the driver removing fuel is more likely to bring down temperatures at #8 quicker than the fan will.


Thoughts?
 
Is there a way to command max fan engagement at a certain temp or will it only sense one temp location and say slow speed at 200F and max speed at 220F (just number examples).

AK's fan I think was sort of commanded in an over ride sort of way as I understood it. If you supplied 12V power and ground a wire it would engage to max. It was not clear to me if you PWM another wire or his 12V power wire was mimicking 100% PWM signal commanding max engagement. Or the ground was completing the PWM circuit and mimicking 100% PWM.

I think anywhere in the engine is better than current factory setup of measuring average stack air temp mechanically after the radiator as it has to be adjusted to come on earlier and its more susceptible to errors due to ambient and air flow etc. Just have to get temperature set point to mimic the temperature required to engage like Kennedy type fan clutch. Its just a faster and "cleaner or sharper" set point from a coolant measurement set point.
 
I don't know but was thinking it could work something like this diagram.
 

Attachments

  • fan-control-idea.pdf
    132.2 KB · Views: 7
schiker said:
AK's fan I think was sort of commanded in an over ride sort of way as I understood it. If you supplied 12V power and ground a wire it would engage to max.
Click to expand...

AK's controller is an adjustable On / Off unit that IIRC is compatible with PWM even though Derale does not list it with their PWM controllers. And Yes, it allows for manual override.

schiker said:
Is there a way to command max fan engagement at a certain temp or will it only sense one temp location and say slow speed at 200F and max speed at 220F (just number examples).
Click to expand...


Yes, am seeing a PWM controller from Derale (#16795) and Hayden (#3655) that will ramp-up speed based on a desirable temperature (adjustable set-point). Derale's literature looks like the controller uses infinite adjustment from 0 - 100%. Hayden's literature looks like the controller uses infinite adjustment from 0 - 100% where 100% occurs at 10F above the manually adjustible 'set point'. I sent requests to both Derale and Hayden for better descriptions of how the PWM controllers actuate the fan through the temperature range.

Regarding controllers accepting multiple temperature inputs, that diagram might work and was along the lines of what I thought about this morning. So far am only seeing units that accept temperature from one input (location) where they also usually allow for manual engagement (dash switch, A/C clutch, etc). In theory, the controller could work from two temperature inputs (locations) by:
> Wiring a ECT sending unit (resistance based) to the controller's temperature inputs.
> Wiring a thermal switch (on / off), at a hot spot location, to the controller's manual override.

Only down side I can think of to using a thermal switch at the hot spot is loss of temperature reading at this location as the sending unit is now an on / off switch.

Another multi-location approach is to use two controllers (reading from two different ECT sending units) for the one fan clutch but this will add complexity and possibly need fabrication to keep the controllers from frying each other.
 
The way I am thinking in my diagram the controller is still only using one temperature input.

The safety temperature switch and the in cab override are parallel grounding paths that will over ride the PWM switching on the ground path and force 100% PWM.

Maybe tap the block off plates on the back of the head for the variable sensor and safety switch. Then use the factory rear head tap for the extra dash guage.

I think that would work fine. You just have to check the safety temp switch can handle the current load of the electro viscous fan solenoid. If it needs more amps then you would have to rig in a relay.
 
I may have gotten too conservative. Probably do not need 2nd safety thermal set to force full speed. Since engines live fine with early engagement mechanical viscous fans. Just need to fine tune set point and fan full speed to be on at about 215f avg head temp. Although I would like to be able to force lock up since it seems to be easy as having a parallel ground.

I saw some thermostatic switches probes used to control Horton type fans. Had options for no switch to close and would reopen a few degrees cooler.

Keep us posted as you progress?
 
Will post back when the project starts coming together.

Am likely to make a two step approach of getting all the sending units in place and then watching the measurements. Final configuration will probably have four sending units (3 in the block and 1 in the hot side radiator hose). After I get all the sending units in place and can monitor the temperature patterns, will decide where to put the fan's input.

And at this time have yet to get a reply from either Derale or Hayden about how their PWM controllers engage the fan. If I get a response, will pass it along.
 
The discussion lacks the built in time delays.
Couple things to think about:
From going full throttle it takes time to open the thermostat(s) up. Then it takes time to heat the radiator up. Last it takes time to get anything but an on/off fan on. The EV or other silicone fluid fan clutch takes time to lock up after being turned on. Obsolete Spring Thermal Fan clutches take time to heat the spring and then move the working fluid before locking up. Worst case is 2 min to kick out on a cold morning before the fan is declared 'dead'.

You can bypass this time delay by triggering the fan before the thermostat opens up so it comes on as the t-stats are open and radiator is hot - IE overcoming the delays by predicting when cooling will be needed. Load and throttle changes may have the fan on with no need for it. This is the challenge with the time delay to turn the fan on.

The worst case actually isn't WOT max RPM. The worst case is heavy throttle, towing, up a slight grade that doesn't make a downshift. WOT at RPM wouldn't get to 210 for me. The lighter throttle would touch 210 because more RPM moves more coolant and allows the fan to kick in faster by moving the working fluid quicker when the thermal spring controlled valve in the fan is open. (The Stock turbo at redline for heavy throttle also contributed to heating.)

The above is stock turbo. Changing exhaust restriction has a major impact on cooling system demand by letting more heat out of the exhaust. Bigger turbo (ATT) means less cooling fan need. Spool Valve on Bigger turbo ruins engine oil because of trapped heat. HX40II also means less than GMx cooling fan use.
 
I think we touched on it and beat around the bush but yeah time is a factor.

I mentioned if you trigger off lower radiator hose as its a big average farther downstream from heat source so to speak. It might be too late as a good safety trigger location.

In my diagram it might also be a nuisance if the fan cools fast and a safety thermostatic switch is "too fine" as in from normally open to closing at high temperature and the reset open temperature is too close especially in a hot location and you don't have enough averaging temperature the fan would cycle. Kind of what thefermanator was saying why ecm uses best average temp of engine.

I was kind of wondering if using a second safety trigger some conditions if a hot location point might trigger max engagement quicker than a progressive controller in a more average local with quick load onset. Maybe if fan was cooling enough it might eventually catch up and kick off max and let the progressive controller take over. I guess depends on trigger temps ("both directions"), ambient, and load/fuel/terrain conditions etc.
 
You must log in or register to reply here.
Back
Top