Home » Public Forums » GMCnet » Mercedes Cooling Fan temperature settings
| Mercedes Cooling Fan temperature settings [message #352803] |
Mon, 16 March 2020 14:10  |
captjack
Messages: 271
Registered: February 2010
Location: Sebastopol, California
Karma: 1
|
Senior Member |
|
|
Since the previous post has now extended to 2 pages, I'm starting a new post with a focus on fan temperature control.
I've installed a simple PID temperature controller to turn on the fan and provide the option of varying the turn on and off temperature settings easily. The output of the controller is a pulse width modulated signal based on how far the temperature is from the set temperature, not just a hard "on" or "off" over a wide range. Based on my intuition and suggestions from the others in the previous posts, I set my unit with a target cold shutoff temperature of 160 degrees. I then set off on a trip to the GMC 49ers rally in Sacramento. The weather was hot and I was running the air conditioner.
Per the instructions that come with the kit, I had installed my temperature probe on the passenger side midway between the top and bottom the the radiator and alongside of the temperature switch provided in the kit which, according to the kit, is set to turn on at 210 and off at 195.
What I found was, that with the AC running, when the temperature at my probe rose to 130 degrees, the water temperature reported from the engine output to the radiator was over 220 degrees. I'm not sure what it would have been if my setting of 160 degrees was reached as I turned on the fan manually which quickly brought the temperature at the probe's location down to 100 degrees and the engine water outlet temp to 195.
What this says to me is that the temperature switch in the kit is inappropriate for this location as it would never turn on....well until after bad things happened. At its current setting it would probably best be installed where the water enters the radiator.
For my drive home, I set the target shutoff of the controller at 115 thinking I could see how the fan behaved when being switched on and off at a rate inversely proportional to the temperature over 115. Unfortunately we were having rain and it was cold, so no AC, and the radiator output never reached 115. So more testing needed.
I did come to the conclusion that the fan should be sent a pulse with varied "on" signal until the temperature at the probe drops to around 120. Also, based on how the temperature at the probe varied with the AC on and off, I can see why modern cars with electric fans are designed to have them turn them on at some level when the AC is running.
Jack Christensen - K6ROW,
'76 Glenbrook/Clasco - "The Silver Bullet",
Sebastopol, CA
|
|
|
|
|
|
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352811 is a reply to message #352803] |
Mon, 16 March 2020 17:21  |
JohnL455
Messages: 4447
Registered: October 2006
Location: Woodstock, IL
Karma: 12
|
Senior Member |
|
|
In the automotive world ( the way it's done) the electric fan On temp needs to be at LEAST 10F over the stat temp rating. Usually a bit more because there is usually a 10F spread in the on to off range to prevent short cycling (hysteresis). This is using engine coolant out temp for ref. A separate issue is AC. Some turn the fan on with each compressor turn on. Some turn it on low with compressor (low meaning low speed via resistor or low by 1 of 2 fans) then to high when over a certain hi side pressure or engine temp is reached. What is different here is it takes time for the clutch to move fluid to the non working chamber, so I would think the Off temp could be closer to the On temp with no hysteresis occurring as fan will continue to run for roughy 2 minutes after Off command causing a built in overshoot. Depending how much air is moved when clutch is disengaged, you may need to add the AC clutch command into the engagement map to give satisfactory AC and head pressures
John Lebetski
Woodstock, IL
77 Eleganza II
[Updated on: Mon, 16 March 2020 17:24] Report message to a moderator |
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352814 is a reply to message #352811] |
Mon, 16 March 2020 18:01 |
Bill Van Vlack
Messages: 419
Registered: September 2015
Location: Guemes Island, Washington
Karma: 14
|
Senior Member |
|
|
John,
What about installing a 180 deg thermostat and setting the PID fan controller to control at 195 degrees?
Bill Van Vlack
'76 Royale; Guemes Island, Washington; Twin bed, full (DS) side bath, Brazilian Redwood counter and settee tops,455, 6KW generator; new owner a/o mid November 2015.
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352818 is a reply to message #352803] |
Mon, 16 March 2020 20:51 |
JohnL455
Messages: 4447
Registered: October 2006
Location: Woodstock, IL
Karma: 12
|
Senior Member |
|
|
The stat only controls minimum temp. Setup might work in warm weather, but will run too cold "on the stat" when cooling conditions are better. But 180 is considered an "alternate temp" stock stat on some GM. (Unlike the 160 the Jegs guys go for). The lower the stat temp, the more potential thermal swing with different conditions. In theory temp swings are bad for gaskets as metals shift. . Probably won't hurt but could keep oil too cold in cold weather to boil water out of oil. I'm guessing these clutches are more go/no go than infinitely variable.
John Lebetski
Woodstock, IL
77 Eleganza II
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352822 is a reply to message #352803] |
Mon, 16 March 2020 22:04 |
captjack
Messages: 271
Registered: February 2010
Location: Sebastopol, California
Karma: 1
|
Senior Member |
|
|
Jim K wants to use the temp at the engine to control things. Some of the Fuel Injection systems (maybe all) allow that. Specifically the FiTech system has that programmable capability via it's temperature sensor usually located in place of the Christmas Tree which would address Jim's concern. It's strictly on/off with no inherent ability to vary the fan speed, much like the temperature switch provided with the fan kit. It is adjustable, however, unlike the temperature switch in the kit.
Jack Christensen - K6ROW,
'76 Glenbrook/Clasco - "The Silver Bullet",
Sebastopol, CA
|
|
|
|
| Re: [GMCnet] Mercedes Cooling Fan temperature settings [message #352823 is a reply to message #352822] |
Mon, 16 March 2020 23:12 |
James Hupy
Messages: 6806
Registered: May 2010
Karma: -62
|
Senior Member |
|
|
I have been watching this thread with a curious intent. Not sure why it's
getting all this attention. The fan has evolved through the years. Started
out being driven by the water pump, and only had a couple of blades. Some
of the very early ones were used on engines that were not even equipped
with water pumps, and were known as thermo-syphon systems. As the engine
heated the coolant, it would rise to the highest point, pulling the cooler
water along with it. That is how the coolant circulates. O.K. for the low
compression, low speed, low horsepower engines of the day. Not very
efficient. Crank driven fans helped with cooling chores. Through the
decades, water pumps, honeycomb radiators, copper and brass materials were
used. Pressurized radiators raised the boiling point of the coolant, and
improved circulation as a result. Along the way, more blades were added to
the fans, along with shrouds and deflectors.
Pretty soon, designers realized that those multi-bladed fans robbed a
bunch of power from the engine, and were very noisy to boot. Thermostats
entered somewhere along the way. Some bright person came up with a self
controlled viscous coupling ( clutch fan) that defied all laws of fluid
dynamics. The fluid was very liquid at cooler temperatures, and became more
thick with a rise in temperature. Unique among fluids. No moving parts.
Bonehead simple. Worked quite well.
So now, we are trying to re-invent the wheel. A difficult undertaking?
Fools errand? Heck, I don't know and am reluctant to speculate on which it
is. It is a tall order in any event. Guess I'll just wait n see.
Jim Hupy
Salem, Oregon
On Mon, Mar 16, 2020, 8:05 PM Jack Christensen via Gmclist <
gmclist@list.gmcnet.org> wrote:
> Jim K wants to use the temp at the engine to control things. Some of the
> Fuel Injection systems (maybe all) allow that. Specifically the FiTech
> system has that programmable capability via it's temperature sensor
> usually located in place of the Christmas Tree which would address Jim's
> concern.
> It's strictly on/off with no inherent ability to vary the fan speed, much
> like the temperature switch provided with the fan kit. It is adjustable,
> however, unlike the temperature switch in the kit.
> --
> Jack Christensen - K6ROW,
> '76 Glenbrook/Clasco - "The Silver Bullet",
> Sebastopol, CA
>
> _______________________________________________
> GMCnet mailing list
> Unsubscribe or Change List Options:
> http://list.gmcnet.org/mailman/listinfo/gmclist_list.gmcnet.org
>
_______________________________________________
GMCnet mailing list
Unsubscribe or Change List Options:
http://list.gmcnet.org/mailman/listinfo/gmclist_list.gmcnet.org
|
|
|
|
| Re: [GMCnet] Mercedes Cooling Fan temperature settings [message #352824 is a reply to message #352823] |
Mon, 16 March 2020 23:39 |
Bill Van Vlack
Messages: 419
Registered: September 2015
Location: Guemes Island, Washington
Karma: 14
|
Senior Member |
|
|
Quote:I'm guessing these clutches are more go/no go than infinitely variable.
JohnL... It sounds like setting the setpoint around 210 to make sure you are controlling the temperature of circulating coolant makes more sense. Jack's system uses PID control with PWM that controls around a setpoint so will modulate to maintain that setpoint.
I understand that this type of fan and control method is used in contemporary vehicles, Mercedes for example.
Bill Van Vlack
'76 Royale; Guemes Island, Washington; Twin bed, full (DS) side bath, Brazilian Redwood counter and settee tops,455, 6KW generator; new owner a/o mid November 2015.
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352825 is a reply to message #352803] |
Tue, 17 March 2020 03:42 |
Ken Burton
Messages: 10030
Registered: January 2004
Location: Hebron, Indiana
Karma: 10
|
Senior Member |
|
|
Let's try this again. The thermostat is a cheap and marginally accurate device. For this reason I usually put the original and the new thermostat in a pan of hot water and heat it. I observe the open and close operation while reading the the water temperature. The thermostat does NOT magically open fully at the desired temp. It starts to open at around 180 or 195 and SLOWLY opens wider if the temp continues to rise. Mean time whether open or closed the water pump is circulating the coolant throughout the engine to keep the entire engine at the same temperature. That is what the bypass hose does near the thermostat and to some extent the cabin heater does too.
When the thermostat is open a limited amount of hot coolant above 180 or 195 is sent from the engine to the radiator. At the time, the same amount of cooler coolant is returned to the engine to be mixed in with the hotter coolant already there. How much coolant is exchanged is determined by the thermostat.
So the question now is how does that replacement coolant get cool enough to keep the engine at around 195 or 180? Well that is determined by the amount of and temperature of the air going through the radiator. We really do not care what temperature the coolant is as it returns to the engine, EXCEPT when is is not low enough to mix in and lower the engine temperature to the desired 180 or 195.
Notice that we have not even mentioned the fan at this point. It can be zero degrees or 100 degrees outside and the coolant can be running at a usable temperature. The thermostat will determine how much cooler coolant will be needed to keep the engine temperature at the desired 180 or 195 degrees.
Only when the returned coolant does not have the capacity to cool the engine sufficiently does the fan get involved. So I still maintain that you determine the maximum temperature that you want the return to the engine coolant to be at and set the fan turn on point there or maybe 5 degrees lower. You only want the fan running when it is necessary and you do not want the slowness or variable temperatures that the thermostat runs at affecting the on again / off again cycling of the fan.
You can have the soap box back now.
Ken Burton - N9KB
76 Palm Beach
Hebron, Indiana
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352826 is a reply to message #352803] |
Tue, 17 March 2020 08:26 |
jhbridges
Messages: 8412
Registered: May 2011
Location: Braselton ga
Karma: -74
|
Senior Member |
|
|
One thing to remember when you use a pulse train to operate the valve in the fan clutch. Since the fluid is fairly viscous the valve isn't going to snap open and closed as a solenoid would. It will essentially integrate the pulses and open partially, the amount of opening being relative to the pulse width. You could consider the setup with an on - off temp switch as pulse controlled with very long pulses. There's a range on the M-B fans which will turn it from about ten percent of shaft revolution to about 90 percent of revolution. Correlate this range to either engine temp or radiator outlet trem and you can use the amount of air flow needed without overdoing it.
--johnny
Foolish Carriage, 76 26' Eleganza(?) with beaucoup mods and add - ons.
Braselton, Ga.
I forgive them all, save those who hurt the dogs. They must answer to me in hell
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352827 is a reply to message #352803] |
Tue, 17 March 2020 08:27 |
jhbridges
Messages: 8412
Registered: May 2011
Location: Braselton ga
Karma: -74
|
Senior Member |
|
|
If you use an analog meter, you can see this action if the pulse repetition rate is fast enough. The needle will fluctuate around the average voltage.
--johnny
Foolish Carriage, 76 26' Eleganza(?) with beaucoup mods and add - ons.
Braselton, Ga.
I forgive them all, save those who hurt the dogs. They must answer to me in hell
|
|
|
|
| Re: [GMCnet] Mercedes Cooling Fan temperature settings [message #352832 is a reply to message #352823] |
Tue, 17 March 2020 09:54 |
 |
Matt Colie
Messages: 8547
Registered: March 2007
Location: S.E. Michigan
Karma: 7
|
Senior Member |
|
|
James Hupy wrote on Tue, 17 March 2020 00:12I have been watching this thread with a curious intent. Not sure why it's getting all this attention.
<snip>
Pretty soon, designers realized that those multi-bladed fans robbed a bunch of power from the engine, and were very noisy to boot. Thermostats entered somewhere along the way. Some bright person came up with a self controlled viscous coupling ( clutch fan) that defied all laws of fluid dynamics. The fluid was very liquid at cooler temperatures, and became more thick with a rise in temperature. Unique among fluids. No moving parts. Bonehead simple. Worked quite well.
So now, we are trying to re-invent the wheel. A difficult undertaking? Fools errand? Heck, I don't know and am reluctant to speculate on which it is. It is a tall order in any event. Guess I'll just wait n see.
Jim Hupy
Salem, Oregon
Jim,
As usual, your observations are effective. But you seem to have missed two things....
The temperature controlled fans were installed for Emissions First and CAFE Second. (The five bladed non-harmonic fan was a noise issue.)
The real reason that the new fan control is of interest is that (like the tuning of clutch springs) making these work seems to be a lost art.
When I have Chaumière sitting on tires again, I plan to do more experiments with the fan control mode and set points, but right now, that is nowhere near the top of my task list.
Oh, and KenB, I am sure that a few more now understand more than they did just a little while ago.
Matt
Matt & Mary Colie - Chaumière -'73 Glacier 23 - Members GMCMI, GMCGL, GMCES
Electronically Controlled Quiet Engine Cooling Fan with OE Rear Drum Brakes with Applied Control Arms
SE Michigan - Near DTW - Twixt A2 and Detroit
|
|
|
|
| Re: [GMCnet] Mercedes Cooling Fan temperature settings [message #352833 is a reply to message #352832] |
Tue, 17 March 2020 10:37 |
James Hupy
Messages: 6806
Registered: May 2010
Karma: -62
|
Senior Member |
|
|
Matt, I didn't mean to imply that high interest in cooling fans was
misplaced use of research and resources that might be better applied to
some other cause. Far from it. I am among those who are quite interested in
where this is headed. My 2006 Chevrolet Trailblazer with the Vortec inline
6 cylinder has an old school viscous clutch fan, engine driven.
However, all my other vehicles have computer regulated, electric motor
driven fans. My Chrysler PT Dream Cruiser is a modern marvel. 4 cylinder,
dual overhead camshaft, 4 valves per cylinder, Turbo Supercharged,
Intercooled, fuel injected engine displacing 2.4 Liters. It puts out 220
SAE net horsepower. The car weighs about 2400 pounds wet. I call these
kinds of cars, "pocket rockets". The computer manages everything from cabin
temperature to transmission shift points to boost pressure, to cooling fan
tip in points as well as speed. All designed as part of the package. And it
works very well. Hardly ever have to even open the hood. Can't get my big
hands in there to fix anything, even if I had too. So complex, old school
diagnostics no longer apply.
Applying this level of technology to a GMC is a tall order. Many of
our coach owners cannot even check their own transmission fluid level
correctly. I am not a luddite by any means. High tech has it's place. I am
just standing back a couple of steps and observing with great interest.
Hope something useful to all of us comes from all this.
Jim Hupy
Salem, Oregon
On Tue, Mar 17, 2020, 7:55 AM Matt Colie via Gmclist <
gmclist@list.gmcnet.org> wrote:
> James Hupy wrote on Tue, 17 March 2020 00:12
>> I have been watching this thread with a curious intent. Not sure why
> it's getting all this attention.
>>
>> Pretty soon, designers realized that those multi-bladed fans robbed
> a bunch of power from the engine, and were very noisy to boot.
>> Thermostats entered somewhere along the way. Some bright person came up
> with a self controlled viscous coupling ( clutch fan) that defied all laws
>> of fluid dynamics. The fluid was very liquid at cooler temperatures, and
> became more thick with a rise in temperature. Unique among fluids. No
>> moving parts. Bonehead simple. Worked quite well.
>> So now, we are trying to re-invent the wheel. A difficult
> undertaking? Fools errand? Heck, I don't know and am reluctant to speculate
> on
>> which it is. It is a tall order in any event. Guess I'll just wait n
> see.
>> Jim Hupy
>> Salem, Oregon
>
> Jim,
>
> As usual, your observations are effective. But you seem to have missed
> two things....
> The temperature controlled fans were installed for Emissions First and
> CAFE Second. (The five bladed non-harmonic fan was a noise issue.)
> The real reason that the new fan control is of interest is that (like the
> tuning of clutch springs) making these work seems to be a lost art.
>
> When I have Chaumière sitting on tires again, I plan to do more
> experiments with the fan control mode and set points, but right now, that
> is nowhere
> near the top of my task list.
>
> Oh, and KenB, I am sure that a few more now understand more than they did
> just a little while ago.
>
> Matt
> --
> Matt & Mary Colie - '73 Glacier 23 - Members GMCMI, GMCGL, GMCES
> Electronically Controlled Quiet Engine Cooling Fan
> OE Rear Drum Brakes with Applied Control Arms
> SE Michigan - Twixt A2 and Detroit
>
> _______________________________________________
> GMCnet mailing list
> Unsubscribe or Change List Options:
> http://list.gmcnet.org/mailman/listinfo/gmclist_list.gmcnet.org
>
_______________________________________________
GMCnet mailing list
Unsubscribe or Change List Options:
http://list.gmcnet.org/mailman/listinfo/gmclist_list.gmcnet.org
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352835 is a reply to message #352803] |
Tue, 17 March 2020 12:06 |
winter
Messages: 247
Registered: September 2007
Location: MPLS MN
Karma: 0
|
Senior Member |
|
|
The early Trailblazer fans were viscous clutch but were controlled by a PWM pulse from the computer. After replacing several of those for many hundreds of dollars, I've since swapped over to the old school viscous clutch used on the newer trailblazers. Its simple but it doesn't seem to fail as easily. I went through at least three of the PWM fans in the first 200K miles. I've since put another 160K on the old school fan with no more issues. I just learned to ignore the service engine light. The controller doesn't like the fact it doesn't see signal from the fan.
Jerrod Winter
1977 Palm Beach
Green Jelly Bean
Twin Cities, Minnesota
|
|
|
|
| Re: [GMCnet] Mercedes Cooling Fan temperature settings [message #352837 is a reply to message #352833] |
Tue, 17 March 2020 13:10 |
Bill Van Vlack
Messages: 419
Registered: September 2015
Location: Guemes Island, Washington
Karma: 14
|
Senior Member |
|
|
Ken,
Thanks for the great explanation... So basically, the thermostat tries to keep the engine water temp at 195 (or 180) by letting cooled water enter the engine. Once that doesn't work, then something tells another part of the cooling system (the fan clutch) to make the entering water colder. In my poor mind, the goal is still to keep the engine water at 195.
PID control of the fan (vs on/off control) is attractive because it can control the fan just enough (engage just enough) to maintain a temperature and no more under all operating conditions so that fan noise and temperature excursions are minimized. We know what the engine water temperature should be, but we don't know what the entering water temp should be under all conditions, so it seems to me that it should control engine water temperature. Just my opinion.
The Proportional part or PID control works like the thermostat in that the more the temperature rises above 195 degrees, the faster the fan runs. The Integral and Derivative parts work to prevent overrun and hunting so that it gets to steady state as quickly as possible.
Engine speed (when the fan is engaged) and speed down the road also affects the cooling performance of the radiator. Hopefully not enough to prevent the PID controller from doing its job.
Johnny,
Quote:One thing to remember when you use a pulse train to operate the valve in the fan clutch. Since the fluid is fairly viscous the valve isn't going to snap open and closed as a solenoid would.
Is the fluid metered because the valve only moves part way during the pulse, or because the valve fully opens during the pulse but only part of the viscous fluid gets through the valve (proportional to the pulse width)?
Bill Van Vlack
'76 Royale; Guemes Island, Washington; Twin bed, full (DS) side bath, Brazilian Redwood counter and settee tops,455, 6KW generator; new owner a/o mid November 2015.
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352845 is a reply to message #352803] |
Tue, 17 March 2020 16:12 |
captjack
Messages: 271
Registered: February 2010
Location: Sebastopol, California
Karma: 1
|
Senior Member |
|
|
Bill,
That's a good question. I'm wondering myself after some quick partial tests while moving. I'm been expecting that the valve opens fully for a short time based on the pulse width, but it doesn't seem to take a very long pulse for the fan to come on fully. There may be some additional filtering needed, or perhaps a dc voltage proportional to the pulse width. The experiment continues...
Jack Christensen - K6ROW,
'76 Glenbrook/Clasco - "The Silver Bullet",
Sebastopol, CA
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352869 is a reply to message #352803] |
Wed, 18 March 2020 07:42 |
jhbridges
Messages: 8412
Registered: May 2011
Location: Braselton ga
Karma: -74
|
Senior Member |
|
|
It will move part way. This because it's moving in a viscous fluid. This assumes the pulse repetition rate is greater than about ten times the opening rate.
--johnny
Foolish Carriage, 76 26' Eleganza(?) with beaucoup mods and add - ons.
Braselton, Ga.
I forgive them all, save those who hurt the dogs. They must answer to me in hell
|
|
|
|
| Re: Mercedes Cooling Fan temperature settings [message #352881 is a reply to message #352869] |
Wed, 18 March 2020 12:27 |
Bill Van Vlack
Messages: 419
Registered: September 2015
Location: Guemes Island, Washington
Karma: 14
|
Senior Member |
|
|
Johnny Bridges wrote on Wed, 18 March 2020 07:42It will move part way. This because it's moving in a viscous fluid. This assumes the pulse repetition rate is greater than about ten times the opening rate.
--johnny
Thanks, Johnny..
I assume that would be a 'position control' solenoid as described here.
Jack,
The info in the link above seems to indicate that the PWM frequency has to be quite a bit higher than one pulse every two seconds. Assuming the fan clutch solenoid is one or the other of these.
Bill Van Vlack
'76 Royale; Guemes Island, Washington; Twin bed, full (DS) side bath, Brazilian Redwood counter and settee tops,455, 6KW generator; new owner a/o mid November 2015.
|
|
|
|
Goto Forum:
Current Time: Fri Sep 06 11:54:28 CDT 2024
Total time taken to generate the page: 0.03608 seconds
|
GMCforum
Search
Help
Register
Login
Home
