after three seasons of driving without a coolant temp sensor, I've aquired a relocation fitting and ready to put it back online.
I didn't remove it before, but I assume I need some sort of thread tape. what should I use?
You should not need thread tape, and as the sensor grounds thru the threads it could possibly interfere with the function of the sensor if it can't make a good ground.
If the sensor is electrical, follow Dave's recommendation.
If the sensor is a mechanical capillary to gauge unit, then you may use thread tape to help ensure seal along the threads. I use standard teflon tape for that.
And if it's electrical and leaks, you can use silicone on the last few threads. Just needs metal on metal contact somewhere.
thanks for all of the input. I am actually re-installing the sensor via a "Mini Sport Alloy Heater Valve Blanking Plate with Water Temp Sender take off"
it seems to have good finger-tight tolerances, but it'll definitely need a seal of some sort. How well it will ground is yet to be seen...
Being alloy and all coated like it is, you may need to run a ground wire from the sensor to one of the studs you mount that sexy plate to.
I haven't tested it to be sure, but I don't think anodizing like that affects the grounding properties.
https://www.google.com/search?q=is+anodizie+coating+an+electrcal+insulator&oq=is+anodizie+coating+an+electrcal+insulator&aqs=chrome..69i57.45368j0j7&sourceid=chrome&ie=UTF-8 (https://www.google.com/search?q=is+anodizie+coating+an+electrcal+insulator&oq=is+anodizie+coating+an+electrcal+insulator&aqs=chrome..69i57.45368j0j7&sourceid=chrome&ie=UTF-8)
Anodizing on aluminum is aluminum oxide and generally is non-conductive per above link.
I would think a simple continuity check with a meter should suffice.
It does not look like the sender will be into the coolant flow very much at all. The usual blanking plate for the heater take-off is about half the thickness of the base of that adapter.
Seeing how short that sender is compared to the mechanical (capillary) one I installed on the Moke, the electrical one would barely be into the flow under the thermostat whereas mine is almost all the way across and beside the #1 combustion chamber. That may also explain why mine seems to run hotter as it is further in and picking up more of the chamber heat versus the coolant temperature.
Quote from: MPlayle on June 02, 2018, 02:17:27 PM
It does not look like the sender will be into the coolant flow very much at all. The usual blanking plate for the heater take-off is about half the thickness of the base of that adapter.
Seeing how short that sender is compared to the mechanical (capillary) one I installed on the Moke, the electrical one would barely be into the flow under the thermostat whereas mine is almost all the way across and beside the #1 combustion chamber. That may also explain why mine seems to run hotter as it is further in and picking up more of the chamber heat versus the coolant temperature.
Wonder if you can find a spacer.
The sensor seat and plate seat should seal if taper is the same without sealant. Probably an AN fitting.
thanks for all the input. this piece is something I discovered that might solve my "nowhere to install it without drilling into something " problem. It was only $12 or so, even for sexy red anodized aluminum ;D. looking to install later today, hopefully I'll be happy with the results.
Anodizing is non-conductive so you will need to ground that somehow. The bolts and sensor threads may dig in enough to break the anodize and ground.
On electric sensors, we always use some form of liquid teflon it will seal yet displace will enough in the tight spots to allow a ground.
You may be able to just sand off some of the anodize it's very thin. Does the sensor bottom out on the taper part? if so just sand the anodize off that area.
Sorry Jeff, taper should be the sealing surface and sanding off anodize will make it leak. I couldn't find any reference that each one was an AN Standard fitting so maybe they are odd pairing.
I'd just scrape off the anodizing under the sensor 'hex', fit a thin copper washer and scrape off the anodizing under the washers/nuts and use heater tap studs as my ground.
right about the taper sealing, I should have said carefully sand off just turning a piece of 220 grit or finer by hand is all it should take maybe even just a red scotch bright pad.
The copper washer idea would work provided it tightens down far enough for contact then it would be the seal rather than the taper but it should work.
Why not just install it and measure resistance between a clean piece of iron (inside the rocker box for example) and the spade terminal, and then between the spade and the sensor body if within a 5% percent, it's fine. I suspect that the tapered thread of the sensor will easily remove the anodizing enough to get continuity. You may not get great continuity between the bling and the head either, but using spring lock washers under the bolt head and chasing out the 1/4-28 threads in the head will take care of that.
Why wouldn't it work? did they think the readings would be higher (and therefore inaccurate) than if you took it right at the thermostat?
I'm with Dave as to using the heater port as an alternate temp sender location - why would it not work? Yes, the #4 cylinder is supposed to be the "hot spot", but if the coolant is flowing well, it should read about the same as for the usual factory location as long as the sensor tip is into the flow.
Mine reads a bit high as I know it sticks in further than the factory and is very close to the dome of the #1 combustion chamber. Some capillary sensors are longer than others, almost all seem to be longer than the factory electric.
As for the coolant acting as the ground medium: when I converted Flur from SPI to carb and added a sensor for restoring the operation of the factory gauge, it was recommended to add a grounding wire and not rely solely on the fluid for a ground. (Granted, the sensor in that conversion ended up in a brass 'T' fitting in one of the always flowing lines rather than the head.) I would tend to agree that installing the sensor should create sufficient contact in the adapter via the threads and seat. Scuffing some of the anodizing from under the heads of the mounting bolts would then provide contact for the bolts to link the adapter to the head electrically.
My only concern is that the tip of the sensor gets far enough into the head to be in coolant flow versus in a potential stagnant pocket.
Quote from: bikewiz on June 06, 2018, 03:37:07 PM
Correct me if I'm wrong but the sender will be touching the coolant, which is a conductor, which is in contact with the cast iron block. I think everyone is overthinking this.....
By the way I made the same piece for my car 6 years ago to run a capillary gauge and on another unnamed forum (MM) everyone poo poo'd my idea, "It's on the hot end of the block it will never work!" now I see DSN and Minisport make them. Should have gone into the business could have sold at least a dozen.
The coolant, while it is a conductor, is not an adequate conductor for a sensor ground it will not carry more than a few ma of current and is a big resistor as well. Just stick some test leads in a cup of coolant on the ohms scale resistance will be in the millions.
Also as a side note, you really do not want to be turning your cooling system into a battery which is what happens when electricity is introduced into a system with dissimilar metals and an electrolyte (coolant in this case) it causes rapid corrosion through electrolysis of cooling system components even the block minis already suffer from this without help.
Turbodave's suggestion is what I would try first as I suggested in my first reply but sometimes that anodizing can be tough to break through in that case a light sand will get it done.
wow, what a great discussion I've sparked!
I'm sorry that I haven't had time to actually install this piece. stay tuned.
Finally installed the alloy fitting and sensor yesterday. All I did was use a little thread tape and bolted it in. Works like a charm, good to put my mind at ease after 3 seasons without it.
On the topic of the "cooler end" of the head, I did notice the gauge sort of hangs between C and N, when before it normally would hang at N (but that was when I had the old radiator with rusty coolant)
Glad that worked out with no issues. Regarding the temp, that's what I expect with a chinesium rad conversion. On crazy hot summer days is the only time I'd hit N and slightly above.
Quote from: 94touring on June 11, 2018, 07:10:03 AM
chinesium
Next new word in the Oxford dictionary. 22.gif
Yep, my same experience....which makes me wonder what the heck is going on with Michael Playle's car.....
At this point I think it is the difference in sensor position that makes mine seem hotter.
The electrical sensor has a short tip - maybe 3/4"? The capillary sensor I used has a long tip - about 2". That makes it stick much further into the head and much closer to the exhaust side of the #1 combustion chamber (a higher heat source location).
I would have to take it out for another run and use the temp gun to report accurate numbers, but I seem to recall the temp of the top of the T-stat housing and the top radiator hose being about 20* less than the gauge, but the gauge being accurate for the head temperature right at the where the sensor mounts as well as the temperature at the #1 spark plug.
I realized I still have the old head from the 998 sitting in the garage and it has the factory sender still installed. I took a couple pictures to try to show the difference in where the senders sit to clarify my comments about why I think the Moke appears to run warmer than expected.
The first picture is showing where the factory sender tip reaches in the head - looking down through the thermostat opening.
The second picture, I placed a pen into the opening with the end of the pen about where the tip of the capillary sender sits - right beside the exhaust valve area of the #1 combustion chamber.
By the way, anyone need/want a used 998 head?