Electric Water Pump question

[Dave Bilyk]

Thanks Bob, I wonder what's different about mine, is it the rear head bleeds? They reduced the cruising temperature 27F, and that will be because they bypass the thermostat. They are only small bore but had a dramatic effect. With the thermostat out I didn't think they would be significant, but now I wonder if there is some thermosyphon effect going on. I will need to try blocking them or putting an orifice in to see what happens.
Dave

 

[Dave Bilyk]

Re: Electric Water Pump running too cool

All, I plugged the rear cylinder head bleed outlet with little or no effect, still drops down to 50C 122F on motorway. Holds a steady temp just above the set point of 85C at idle when stationary. I have checked the heater and there is no hot water going through it, not surprising since there is no mech pump differential pressure any more and the heater valve is closed anyway. There are no other circuits, so unless I have missed something, the only answer is the circuit resistance is lower than the pump controller can handle or there is a natural thermosphere going on which seems unlikely. I can try condulting DC. Any other ideas?
Dave

 

[Dave Bilyk]

Update - I contacted Davies Craig, who are very helpful, they suggested that I put thermostat back in with two 5mm holes. I did this, but the effect was marginal. Blanking off a third of the rad got the temperature up to 65 but that was it, and compromised ability to maintain temperature when stationary, just an experiment.
I then tried another thermostat starting with two 1.5mm holes, but no improvement which is very baffling as it seems like the thermostat is leaking or not closing, but I tested it in hot water and there is plenty clearance in the housing so no interference possible.
Anyway, my next step is to get a 3 way radiator bypass valve and install it at the front. :idea: That has to solve this :evil:.

Dave

 

[Brett James-McCall]

Dave, I think I may have some Moroso restrictor plates somewhere that you could have go with

 

[Randy V]

You may want to take another look at my build here to see how I will be integrating a BMW Bypass style thermostat:
GT-Forty.com

 

[Dave Bilyk]

Brett, thanks for the offer, but I have ordered a bypass thermostat and other bits and bobs from Car Builder Solutions. Will arrive today so hopefully have this installed over weekend.
Randy, the bypass thermostat looks very much like yours, so installation will be similar but without the fancy work, at this stage anyway as I want to get up and running.

Dave

 

[Dave Bilyk]

Update, after much head scratching trying to use the hoses I have, I ended up cutting up some hoses meant for the Subaru to get the arrangement to fit. So here it is before final adjustments and hose clips, the angle on the bypass port makes for a contorted arrangement, but lets say it's an art form eh!
Randy, I saw that you cut the angle bit off, that would be much neater but I don't have time just now. I will hopefully complete and test tomorrow.
Dave

 

[Randy V]

We welded expanded hose ends onto the T-Stat housing.. Care was taken to go slowly and not overheat the housing from the welding operations.

 

[Jac Mac]

Randy, that t/stat appears to be one piece/non-serviceable, have you actually ran the car /engine with the t/stat housing welded in that manner to ensure its not affected?

 

[Dave Bilyk]

Randy, great job, yes you would have to be very careful with the welding. As for mine I went for a good test run today, great in traffic, but still over-cooling on the motorway, now at 65deg C about 150F. Now I will try an orifice at the thermostat inlet, there is probably too little resistance in the circuit and so the pump is delivering too much, there may be a thermosyphon effect too. As Brett said Moroso provide various size orifice plates, so these must balance the system resistance to the pump, but I can cut my own anyway. If necessary I can optimize to make sure I maintain cooling when stationary.
Dave

 

[Bill Musarra]

Dave,
I am not running any type of thermostat and on the open road I get temps in the 160-170 range with a 35 gph pump. Do you know for sure the temp that the t-stat closes/opens?? A test may be in order. You may want to consider a pwm to the pump to slow it down. I have one but just haven't put it in yet.

Bill

 

[Randy V]

Randy, that t/stat appears to be one piece/non-serviceable, have you actually ran the car /engine with the t/stat housing welded in that manner to ensure its not affected?

Hi Jac,

You are correct in that they are sold as one piece and non-serviceable..
No I have not run the car yet (engine on the dyno yes).. However, I have tested both thermostats in a kettle of hot - then - boiling water and they both functioned. Were they functioning at the same temp they were originally spec'd for? I don't know.

 

[Randy V]

Dave, I have to say that in all the years I have been here on GT40's, Your car is the very first GT40 that was having trouble building heat that I can recall...

I thought one of the key selling points of the Davies Craig units were that they operated at variable speeds to control the flow and manage the engine temps. It is very odd that this does not work for you.

Have you double checked your guage against a known good mechanical guage?

 

[Dave Bilyk]

Thanks for all your help guys, it's appreciated!
Bill, Bob Putnam said the same, and this is very baffling. Is there a restrictor in your water circuit?
Thermostat working
This thermostat does work, tested on the cooker with no temperature measurement. In the car, at idle, the radiator was cold as the bypass circuit warmed up, then finally the radiator was hot with fans on, maintaining a set point of 85degC 185F. Agreed I might do some further testing to establish good data.
Gauge check
To answer your good point Randy (it's baffling the hell out of me too!), at this point, the 3 measurements (controller, block gauge and my own gauge on the thermostat inlet) agree, and also when I get on the motorway, the block gauge and controller both show low temperatures. The controller works and controls the pump as advertised from my observations during warmup, and all controller indications are consistent with this.
Bill pwm
The Davies Craig controller works like this, up to 20C 36F below set point, the controller pulses the pump 10sec on 30sec off. from there to 5C 9F below set point 10sec on 10sec off. Then from there the pump runs at 6V, ramping up to full volts at set point. Fans on at 3C 5F above set point. All controller indications are consistent with this.

I have some thermocouples so at this point to understand fully, I will take some detailed measurements at critical locations.

Dave

 

[Dave Bilyk]

Well, it's been a good and bad day. The bad bit is that I left my pressure cap off, then today I moved the car out of the garage, it fell under the wheel, couldn't find a new one as motor factors closed. :cry:
The good bit is that I instrumented the cooling system with thermocouples which will help understand what is happening. Since I couldn't now road test, I contented myself with a warm-up at idle and confirmed that the thermostat opens at the correct temperature.
I put in 8 T/C's covering rad in and out, engine in and out, thermostat in, pump in, air in and out, so I think I will get enough data to understand what is happening.
Without the pressure cap, I ran until it burped and expelled some water, which happened just after the thermostat operated, so temperatures had not reached steady state, but the thermostat operated on the button between 78 and 82degC so at least we know that.
Once I get a rad cap I will repeat measurements, do a road test with a data recorder (that'll be the wife) and get steady state data. This is now like how to have fun solving problems!

 

[Randy V]

Dave -

Just a thought...
Truckers use a zippered shield in front of the radiators of trucks up in the north country where I am in the winter...
You may want to consider blanking off the top half of the radiator with a sheet of heavy plastic or aluminum to reduce the efficiency of the radiator...

 

[Dave Bilyk]

Update, warmed up with the thermocouples installed, taking readings every minute, then took the car for a run and got reasonable data.

The thermostat is an 80degC model and does open at 80degC according to the thermocouples. On the road from the temperature data around the radiator and pump there seems to be no issue with temperature control, because on the open road, the thermostat temperature maintained between 78 and 82degC. From this new data the temperatures are as expected, and not cool as the dash gauge suggests.

At Idle with fans on, the temperature is controlled by the Davies Craig controller, which switches fans on at 88degC.

What I cannot understand is the data from the engine side.
At idle with fans on, the thermocouple at the Engine OUT position, the Dash reading from the engine block sensor, and the DC controller read the same as the Tstat IN, which seems to prove that calibration is good.

However, on the road, while the Tstat IN temperature is controlled to 78 to 82degC, the Engine OUT and DC readings drop to 75degC at 40mph and 72degC at 70mph, while the Dash reading (engine block) drops to 70degC at 40mph and 65degC at 70mph.
To have an engine temperature which is lower than the Rad IN temperature is impossible, so I don't know why it appears to drop with speed. If it was a measurement issue, why would they read the same at ambient temperature, and at 88degC, but have a significant error in between?
Still thinking about it.

Randy, I had already tried radiator blanking (1/3 of rad area), but it just compromised the cooling at idle.
Dave

 

[Andy]

Hi Dave,
Nice work with those measurements.
I guess you are not using an oilcooler, ignition is enough advanced.

My feeling is the flow of your pump is perhaps too high for the system so the system is chasing its own tail from a thermodynamic perspective or cavity in the pump introducing air bubbles. you can lower the voltage to the pump to reduce the flow and keep it constant flowing slow, see if that helps ? You might be right, there are too many parameters and control loops.
I dont like the idea that you shut the pump for 30 seconds. Why not control the pump linked to the engine rpm as per normal waterpumps ?

The thermocouples are measuring the steel around the water but there is also some lag here between real watertemperature and what you measure. Also different ways of mounting the thermocouples can blur the measurements.

Why not putting the pump before the thermostat ? Perhaps it is sucking now too hard and the under/low pressure affecting the thermostat operation ? Temperature and pressure are also linked...your idle temperature should be between 90 and 100 degrees. So you could go up at idle.
The fans on the radiator are turning in the good direction ?

 

[Bill Musarra]

Dave,
I currently run the cooling pipes unrestricted.Still on the fence as to adding the PWM. Maybe this winter. Don't want to mess around with the KISS principle. I will sacrifice fuel mileage for performance for now. Have my ECU setup for that, and have gotten improved fuel mileage to boot. Planning a full fledged stint on the dyno soon. We did a rolling tune back in the spring. The temps outside were too cold(20 degrees F.) for anything more. Still got 160 in that temp on the highway.

Bill

 

[Denny Hughes]

Davies Craig 150lph pump with header and controller is going in my SLC