OK, new project here and I need your help.
The idea is to replace the existing radiator cap with one that's a temp switch mounted on it, thereby avoiding the need to mount the temp sensor to the radiator tank.
So far it's feasible. In fact I'm in the process of soldering a bung onto a cap to try it out. But we have a Quest to find the perfect switch. Here's my suggestions, I welcome your comments.
A gas-tight connector (O-ring or silicon seal type).
A Dual switch with set temps of ~185 and 200F.
An extended bulb, approximately 2" long below the threads. It could be much longer for crossflow applications.
Prefer an 0-ring seal and straight threads.
Think any of you can help with this?
Jim
ps: the advantage of a radiator mounted switch is that the fans will follow the temp of the radiator and switch off much more quickly when the car is parked.
Incidentally, this is for a blanking cap with no pressure relief. There needs to be a standard cap somewhere else in the system.
Jim,
I know you will disagree here but the important information is not what temp the fluid enters the rad at. The fans should come on based on how the rad is cooling, therefor place it at the bottom of rad or easily in the bottom hose. Been running mine that way for two years and no issue. Going down the highway and up a steep hill, coolant existing motor is hotter than usual but you don't need your fan coming on at HW speeds do we. Bottom of rad is also effected by the fan after shutdown so fan should not run excessively.
Mike
Quote... important information is not what temp the fluid enters the rad at...
Afraid that I'll have to disagree on this one.
Monitoring of engine operating temperature is based solely on max cooling system temp and is regulated by the t'stat.
Fan on/off temps are always set higher to ensure that the t'stat is fully open before the fan engages. In this regard t'stat and fan switch should be linked to the same datum (ie. max temp at the t'stat.)
So, for instance, if you have a 180F stat that is fully open at 195F, then the fan switch setting should be on the order of, say, 205F/on and 195F/off.
I don't think linking fan switch settings to the rad coolant out is best practice.
As far as your idea,Jim, seems to me that finding a home for the switch in the intake manifold or t'stat housing might be just as easy....or you could insert a small piece of pipe in the upper rad hose - something like this but better executed!
bunginupperradpipe.JPG
Graham,
I know your background but.... The largest auto manufacturer in the world is wrong? VW has had their fan switches in the lower part of the rad since 1975 in the first Golf. The fan is a feedback loop. Prevent the coolant from starting out too hot into the motor. Having the fan switch at the t housing or top of rad lets your fan run going down the highway even if the coolant is not to hot after the rad.
I believe Artie has a radiator mounted fan switch and his cycles after shut down on a warm day. It will run a while until the radiator cools sufficiently to open the switch and then come back on a minute or two later. I'm assuming because the coolant flows a bit because of the temperature difference between the radiator and block? Just relaying what I've seen, not sure if this is normal or not.
You know I did think about that Mike and a lot of what you say seems to make sense. Here's why I'm not approaching it that way though: First, I do not want to cut my lower radiator hose. I went to a lot of trouble to get one that fits well and I want to leave it just the way it is. I also don't want to remove the radiator to add a bung to the left side tank. It is a lot of work to remove the radiator from my car. But I do think there might be some merit in using the exit temp. The down side is that engine temps would tend to run hotter unless you can find a lower temp switch, and engine temps would probably not be as stable, as Graham mentions.
Now, in my particular case installing a switch in the intake is an even more difficult option than putting it in the radiator tank, as the wall of the only available spot on the manifold is too thin to support the required threads. (It is a fabricated manifold) and it would have to have a bung welded into it. That is not a practical option at this time since the entire blower/manifold assembly would have to come off to do it. In Dan's case we have found little difference in results by moving the switch from the outlet to the inlet on the engine, both of which are at high locations and isolated from the radiator.
These are sort of opposite scenarios. On my car the fans do not turn on because the switch is mounted in the surge tank and it doesn't get hot enough. On Dan's car the fans do not turn off because of heat soak up high on the engine. The fan can run for 45 minutes and the top of the engine still hasn't cooled down enough to switch off. Obviously battery life is a big issue.
BUT in both cases the radiator inlet tank will be at engine thermostat temperature while running. (Note BL used a fair number of radiator mounted fan switches.) The radiator cap is located on the inlet tank. In Dan's case the fan cooling the radiator will cause the quickest temperature drop after shutdown to be in the radiator, whether at the bottom or the top, resulting in the quickest fan turn off. In my case the temp at the radiator inlet tank will be high enough to turn on the fans. Therefore, the only reason the radiator cap is not a good location for a fan switch is due to the chance of an air bubble around the switch. In fact, provided there is fluid contact this location should be virtually identical to Graham's illustrated hose mount.
That is why I think a switch with a long sensor bulb might work well, and for both of these applications the longer it is the better since they are both cross-flow.
The 185 and 200 degree switch points were a suggestion, it could be a bit higher but I'd really prefer to not go over 195/205.
Now many conversion cars may not have enough room above the radiator cap to use this solution, but for those that can I think it'd be an elegant install.
I have completed the install on my car today, using a Saab switch (20mm, stubby, pigtail) by just soldering a bung to an old block-off cap and fitting a silicone rubber gasket. I'll see how it works tomorrow if it doesn't rain too much.
Jim
That cycling behavior does make sense with a typical fluid layout, but Dan's 1UZ has both fluid lines at the same level on top the engine so thermosiphoning should not occur. That would explain why cooling the radiator does not cool the engine after shutdown. On my car, power is cut to the control circuit on shutdown.
Also while I agree that controlling coolant temp supplied to the engine does make at least some sense, for these two applications at least it is far more practical to use a cap mounted switch if it can be done correctly.
Jim
Jim,
Valid points. For switch I used flex a lite variable controller. Yes I to cut my bottom hose but I was already using a metal pipe by the oil filter to prevent hose issues near the "frame rail".
Hope your new setup solves the issues you are having.
QuoteThe largest auto manufacturer in the world is wrong? VW has had their fan switches in the lower part of the rad since 1975 in the first Golf. The fan is a feedback loop. Prevent the coolant from starting out too hot into the motor. Having the fan switch at the t housing or top of rad lets your fan run going down the highway even if the coolant is not to hot after the rad.
A typical temperature drop across the core should be about 12-15F – it's a fine balancing act to ensure that the coolant resides in the core long enough to allow heat transfer to the air but at the same time allow sufficient coolant flow through the engine to avoid hotspotting.
There's no reason one couldn't use a lower hose fan switch location and spec the fan on/off temp settings accordingly......but, whether the fan switch is installed on the hot side or cold side, the fan switch cycle is normally spec'd above the engine t'stat range.
The bottom line here is that with t'stat fully open, the fan switch will activate when engine coolant temps dictate .
So what I really would like to know is if any of you can remember seeing a dual circuit switch with a long bulb?
Jim
Maybe I'm missing something here. The coolant circulates below the spring loaded poppet valve on the rad cap so the only way to get a sensing bulb into the circulating water is to drill a hole through that poppet valve. If you do that the radiator will not pressurize since it is no longer a sealed system. The spring in the cap is what regulated the max pressure in the system.
Jim S,
From Jim B's first post:
"Incidentally, this is for a blanking cap with no pressure relief. There needs to be a standard cap somewhere else in the system"
Anything here?
http://www.tridon.com.au/products/Tridon/35/483/switches-and-sensors/2014/thermo-fan-switches
Jim,
When you upgrade to MS3 you can just use your ECU to control the cooling fans. There's even a pop-up window for fan control in tuner studio under "basic/load" settings. From there you can program it to come on and off at any temperature(s) desired.
The best part is if you need or want to change the fan switching points, it just takes a few taps on the key board and your done!
Bill
Thanks Graham, I knew I must be missing something.
Graham, unfortunately I was unable to find anything with an extended bulb in a 3 terminal switch. Invariably 2 terminal means a single switch. Good reference though.
Bill, you are right, in fact I should be able to do it now.
Weird thing though. We hooked it up on Dan's MS-II and the controller went haywire and lost essential functions, I think it quit running the ignition advance or something and we had a heck of a time just getting that back so the engine would run again. We decided not to push our luck. I may try it again when I do the MS3 upgrade, probably will in fact. The one on the megajolt controller in the Roadmaster works very well. But in the meantime I want to try and sort this out so I can continue to concentrate on other tuning concerns like the idle and start cycle. Plus if it works out well it'll be a good solution for Dan as well as for other owners. Then too, there is something to be said for isolating systems from each other. Sometimes simpler is still better.
Jim
Jim,
rather than search for a unicorn. Why not "french" a sensor into the cap.
That way you can sink it to any depth that you want. And the range of usable sensors increases dramatically.
It also lets you use a sensor that is compatible with the MS.
If nothing else it would look more refined.
Cheers
Fred
P.S. just to mess with you, I've been playing around with the BMW electric water pumps. They are a small, powerful PWM brush-less pump.
Have one rigged up to a Dodge 12v diesel and another one on a 240Z with a RB26DETT. Gives you pinpoint control over the engine temp, reduces parasitic losses and eliminates the antique t-stat completely. The best part is that in failure mode it goes to 100% duty cycle.
I'm in the 'top' camp.
How about a variable sensor, with a capillary bulb that tucks into the top of the tank via the top hose? Choose your own temperature, and no need to cut anything.
Kenlowe no longer make aftermarket fans, but this is pretty much their system been around maybe 40 years, and on four of my cars.
http://www.demon-tweeks.co.uk/performance/fan-controllers-thermostats/davies-craig-mechanical-thermal-switch-fan-controller
Perhaps not as much fun as breaking new ground.
Ivor
Bill,
Jim and I actually had the fan control in the MS2 working as designed and were pretty happy with it, until we tried to get tach control out of the MS2 box. When we did that everything went to crap. We then spent several hours backtracking and troubleshooting until we finally got the car to run again.
The 1uz has a temperature switch on top of the engine that ran the original hydraulic fans, so those would obviously cut off when the engine turned off. Using that switch, it does not cool down enough until the battery is too low to run the fans.
QuoteUsing that switch, it does not cool down enough until the battery is too low to run the fans.
For that very reason, I've never understood the logic of fan(s) wired as to keep running with the ignition off. I can see having a manual override switch to engage the fan(s) when needed but to have them draining battery after engine off could be problematic.
Slight problem. The crossflow radiators have the fill neck on the suction side. (Of course dummy.) This doesn't work at all. Too bad though, it seemed like a clean solution. Maybe with a much cooler switch point.
Might work with a downflow but only if the fluid can be kept in contact with the bulb.
Well, thanks for the assist guys. Fun, if not successful.
Jim
Seems to me that late model MGB's had the fan sensor stuck in a hole in the upper tank (forward face) and retained with a wire clip. If you're going the soldered bung route, why not solder it into the pressure side end tank, similar to original. Your fan switch needn't be so long and more available, then.
Yes, it is flat on the end and with some heat sink paste could simply be clamped to the tank. I will give it some thought. Maybe a fairly simple "C" shaped clip with a threaded hole and screw in the sensor as the clamp. Could be a good idea. Make it out of about 10 ga. aluminum maybe.
Jim
I have one like Ivor posted tied to my OEM electric fans with the probe stuck way up into the upper hose. This is as a backup to my engine fan.
I have a temperature switch that is on a probe that fits into the radiator fins (air space) not water space.core. It is about 1/3 down from the top and works just as intended. It is easy to experiment by inserting it in different parts of the radiator...
Here's my next attempt:
IMG_0017.JPG
Had to trim the bracket a little to fit it through the radiator mounting frame.
IMG_0018.JPG
You might want to put some heat paste on the sensor where it contacts the rad, make sure it picks up the temp.
https://www.amazon.com/ARCTIC-Compound-Performance-Heatsink-Interface/dp/B0045JCFLY
Ivor
Yeah, I did that but thanks. I need to go back and short the leads to make sure the fans activate. They should but I'm still not getting the right results. More later.
Jim
Have you checked it out with an infrared heat gun Jim?
Probably the easiest way to home in on the sweet spot.
The heat loss around the sensor might be too much as well. Might need some sort of insulation to function correctly.
Cheers
Fred
Turns out the switch was the wrong spec. I'll need to find one that turns on at about 185 and 195F. I think it's a 22 x 1.5mm thread.
Jim
You need the early VW Audi rad fan switch Jim.
Dual range, 95/84 - 102/91 degree C, on /off temps.
22 by 1.5mm thread.
Will run a 450 watt fan without a relay but I always put one in anyway.
Part # 321 459 481A
Cheers
Fred
Thanks Fred (just sent you a PM btw)
If I could get one just a little cooler than that I think i'd like it better.
Jim