I thought I had posted this but I saw a logged out message after I hit "submit" and I don't see it anywhere, so here it is again.

The motor is a SC 3.8L V-6 in a Brunton Lotus 7 kit car. It was out of a 06 Pontiac Grand Prix with less than 50K miles. We've put an additional 10K on it in the kit car. No mods to the engine other than the cooling system fill port described below.

My question has been developing/irritating me for more than a couple of years. It has to do with water temps varying radically while driving in a steady state (low to mid RPMs, speeds fairly constant in the 40-75 range, level on freeway or surface roads, no traffic or significant slowing, manual fan off or on ). The needle will typically go back and forth in a matter of seconds from 210 to 220. And then sometimes it will go down to 190 or up to 230. I have done all the normal things (new radiator, flushing, new OEM drilled 195 thermostat, new temp sensor, and careful filling with distilled water and water wetter at 4 oz/gal and new 16# radiator cap and new adjusted belts). BTW, the temp is pegged at 200 for the first 10-15 minutes of driving and then goes down to 190 for a couple of minutes and then on up to the usual 210-220 range where it jumps around for no apparent reason, again, manual fan on or off makes no difference. ambient temps (55-90F) make little difference.

When I got the kit car from the original builder, it ran 190 to 200 very predictably depending on ambient temps (up to 95F) or traffic (stop and go) or freeway cruising. Temps varied gradually and as expected for the changing conditions. On track (5k+ RPM and 50-120 mph), I soon learned the necessity of painstakingly filling and burping. Even before I learned that, it never went over 235. High variable temps were easily explained by air in the system. So I installed a fill port at the high point, next to and downstream from the drilled thermostat to easily burp the system. I have an had caps of 18# to 25# on the radiator and fill port caps of 16-18# and no leaks anywhere.

I have been told by one reputable GM mechanic that water pumps can (even if rarely) fail gradually. I believe he said a spun bearing. Which I presume is different than spinning a crankcase bearing where the motor seizes. Meaning, I suppose, that the impeller is still turning but irregularly and unpredictably due to friction, and not matching the engine speed. And that circulation is slowed significantly until the impeller moves normally for a bit and moves a bunch of much cooler water from the radiator. He did not suggest the gauge may be bad.

So does anyone have any other suggestions before I change the water pump? Thanks.

I'm not for sure but it'* possible that the OEM temperature sensor is not working properly without the proper coolant mix, meaning it may be inaccurate with your distilled water/water wetter. Also water pumps need the ethylene glycol mixture for proper lubrication of the impeller bearing or at the very least water pump lube which if I remember right comes in about an 8 oz. can. I assume you're not using a traditional coolant mix because it'* prohibited by track rules? Any coolant leaks result in a very slippery track? So if you haven't been using an ethylene glycol coolant mix or water pump lube, yes it'* very possible that water pump is on it'* way out.

thanks Rog.

Red Line says this about water wetter "Rust and corrosion protection allows for use of straight water in racing or reduced antifreeze levels in warm climates". I and everyone I know in my race group (more than a couple dozen Formula Mazdas all with the same spec rotary lump) used Water Wetter/distilled water only; no additives. I only raced for 12 years but some of those guys have been racing more than 25 and that was the standard.

As for the temp sensor, increased temps and variability started being a concern with the old temp sensor. The issue is still qualitatively the same with a new sensor that I put in a few hundred miles ago.

The mechanic mentioned cavitation but my research on that draws a blank But an inspection may tell more. He also said that pumps can wear and lose performance as the impellers can rust or crack. He has seen them spin on the shaft and not turn the proper speed. This was news to me, but rings a bell for my symptoms. The bottom line is that there'* no telling how the cooling system was maintained when it was still in the original car. the kit builder is a racer and pretty smart and thorough guy, so I'm confident he used a decent anti-freeze and/or water wetter. Probably WW, coming from Arizona.

From what I see, pulling the pump is not difficult, so guess that'* the next step.

Thanks.

Does anyone know how a gauge acts when it goes bad......besides just dying with no movement? I am getting push back from a knowledgeable mechanic and the established local radiator shop about the pump being the culprit. The only thing left is the gauge. Anyone seen a gauge go rogue and give faulty readings?

I tend to agree about the water pump, it'* either bad or good. A bad bearing will usually make noise, or at least have a little play in it. Either way, if a bearing is bad, if the shaft is still turning the impeller is still turning . . . and doing do in about the right position unless the bearing is so bad that the shaft can move around so much that coolant would be gushing out anyways. The impeller on the back isn't known for eating itself. In a 50,000-mile donor I bet it'* in perfect shape.

What kind of gauge is it?

What kind of temperature sensor is it?

What are the dimensions of the radiator?

How many rows does the radiator have?

Has this engine been in this car the entire time you've owned it?

How long after you acquired the car did this issue start?

about 3 years

Thank you.

Yeah, that'* a small radiator so I was thinking in terms of a sticky thermostat opening a little late and the radiator takes a bit longer to recover, etc. Sounds like this isn't common.

The rest seem okay enough, except possibly the gauge. I agree, a gauge replacement is probably the next best course to take. If the issue continues after replacing the gauge then I'd suspect the thermostat next. I saw it cot replaced already, but was it replaced as part of this issue?

I've swapped the temp sensor about 4 times in the last 8 years. Not a lot of reliability in readings for the Delco sensor.
Been using the Standard brand sensor for a few years now, so far so good.

yes I did replace the T-stat not long ago with an dealer OEM. not a parts store generic.

The Speed Hut support guy has given me a couple of readings to check under various conditions. 4.9 V to the gauge. Then resistance; should be 108 ohms going down to 82 ohms as the temp goes from 190 to 210. So I'll do that and see what he says.

**** I guess I better stock up.