Jeep XJ -> Cloosed to Open cooled

[sandalscout]

Hey all,

I'm gathering parts to convert my cooling system from closed to open (and eliminating the heater valve and leaking coolant bottle), and am also installing an aftermarket cooling temp gauge. I'm looking for some insight into the temp probe placement.

I found a writeup where someone used a 3/4" tee, and put barb fittings on 2 two sides, and an adapter to plug the probe into the middle section. He spliced this into the middle of the heater hose going to the core. I have assembled one of these, with the intention of doing the same thing, but had another idea.

I am using an adapter (http://www.jegs.com/i/Moroso/710/63730/10002/-1) to add the filler neck and cap into the upper radiator hose. What if I tapped this to fit the probe into the side of it? My initial concerns are that the probe is after the thermostat in both situations (but this is normal, right? too cool = thermo not opening?) and that the neck will be cooler than a true reading do to air sitting around it? if the neck is inline in this area, though, I think most of the air will settle out in the rop of the radiator eventually.

Any thoughts? Appreciate it!

 

[wADVr]

Any reason you want the aftermarket temp guage? If your XJ doesnt have one factory its pretty easy to swap the cluster and then also gain battery voltage and oil pressure. I cant remember for sure but I think its pretty much plug and play as all the wiring should already be in place. The factory sensor location is to the very rear of the valve cover on the manifold side (drivers side) IMO this is a good spot because its a lower flowing area of the cooling system and on top of the cylinders so it theoretically should be one of the hottest areas.
Another option to getting the coupler to mount then filler neck would be to take the radiator to a shop and have them braze a filler neck on-should be less than $30. If you have a plastic tanked radiator that isnt an option but maybe upgrading to a full metal radiator with the filler already in place would be money well spent.

Good luck

 

[baddis]

i snapped my coolant temp sensor off flush with the block when i swapped out my instrument cluster. so far this is the best alternative i've found to drilling and tapping it out.

http://www.deftracing.com/prosport_gauges/water_temperature_sender_radiator_hose_adaptor.htm

maybe this will help you. i've found a couple other manufacturers online that make these. i think i've also seen some on ebay.

 

[sandalscout]

Thanks for the replies!

T&A - the reason for not going to the XJ Gauge cluster is simply availabilty and price. I have found clusters for around $50 plus shipping, and the new senders are about $60 from what I understand. I have next to nothing in my oil pressure and water temp gauges and the mounts for them currently.

I had hoped to use the stock sender location on the intake manifold, but it's not feasible with the size of the thermoprobe, unfortunately. I don't think I could easily tap into the thermostat neck, as I've had suggested as well.

Unfortunately, it's plastic tanked, so the filler neck right onto the radiator is a no go. I seriously considered replacing the radiator, but really wanted to get a temp reading before hand. I THOUGHT I was having an overheating problem as first (smelled coolant leaking from the pressure bottle), but the idiot light has never come on, even after a 3-4 stint of offroading with an XJ that WAS overheating slightly. I'm happy with the gauge I have, so once I get it in there, if I do discover I'm running too hot, I'll start saving money for a 3core all metal radiator and a new water pump. If not, cool, I can spend my money on other gear. I honestly just can't justify dropping $400 on more cooling stuff that may not be needed, especially on a vehicle that's only driven 2 days a week. I'd love to, but can't.

Baddis, thanks for the link, that's essentially what I built, except smaller to fit into the heater core hoses. I think I will probably go ahead and tap the filler neck, so that the probe is in the upper hose (I've read that this is the fastest responding spot for fairly accurate readings, but that I should see about 10-15 degrees below the maximum REAL temperature in the block). Since the hose is flexible, the filler neck will weigh it down just a little bit, allowing the the neck and therefore the thermoprobe to stay below the inlet to the radiator and stay full of water, eliminating false low readings. When I need to fill it up, I'll just lift the hose and make it the high point in the system.

Thanks again!

 

[madizell]

You can determine coolant temperature externally if you need to confirm actual temperature by using a simple and inexpensive IR thermometer available lots of places. Check the temp at the thermostat housing and at the inlet and out let hoses for the radiator, and you will know if the motor is actually overheating, or if the coolant control is attributable to radiator cap (which fail), or something else.

 

[sandalscout]

That's a really good point that I hadn't thought about. I'd still feel more comfortable being able to monitor all the time with the gauge, but I could get the IR thermometer to pinpoint heat, determine if the gauge is off by much, etc.

Thanks, I think I'll grab one. I have wanted one to test tire temps for a while anyway.

 

[ntsqd]

One thing about the IR tools, their reading is skewed by the surface that you point them at. I have seen differences of 5* to 25* A raw cast aluminum surface will read a different temperature than a flat black painted cast aluminum surface, even if they are exactly the same true temperature. At work, where the monitoring can be critical, we paint flat black spots at every location that we're likely to want to know the temperature of.

 

[madizell]

Temperature differences will become apparent when various locations are checked. IR's will not read high due to differing surfaces - usually will read low or not at all. If one temperature is significantly lower than the others and the surface material and finish are different, it could be a read error. Flat black paint will cure that.

You should expect to see a significant difference in one area anyway, as between inlet and outlet hoses where you would want to see perhaps a 30 degree drop. But since both surfaces are rubber hose, any read errors should normalize and at least the shift in temperature should be reliable.

When I see 195 degrees displayed on the digital temperature gauge and the IR thermometer pointed at the sensor location, I figure they are reasonably reliable indications of internal average temperature. Hot spots can be seen across the heads due to coolant distribution, which is normal.

I just find it fascinating to scan temperatures on the engine. Cast iron headers can run 435 to 500 which is a lot of heat, but the stainless steel headers on the V-8 can show off the scale (mine tops at 999) when the engine is loaded, which gives you a good idea of what needs protection under the hood, and where not to route wires and such.

You can use these things around the house as well, to check room temperatures or the temp of that bottle of red wine you are serving to guests.

As with any tool, you need to know its foibles, but I find the IR thermometer to be a great diagnostic tool. I discovered which of the 6 cylinders on the V-6 was not firing all the time just be checking header temps. A lot safer than pulling spark plug leads and only took a few seconds to do it.

 

[robert j. yates]

As with any tool, you need to know its foibles, but I find the IR thermometer to be a great diagnostic tool.

Agreed.....I have used the IR gun extensively on my rig as a diagnostic for water temps, engine oil temps, tranny temps and on the ac condensor. Rather than chasing my tail and guessing, the IR gun pretty much told me what I needed to address. A spritz of flat black paint is very helpful on aluminum surfaces.