Davies Craig Electric Water Pump Gen 3 - with optional hand controller

Car - MK3 1997 GTI

For your information I have typed half of this review with one hand due to a recently fractured collar bone so I hope you find it helpful! I didn’t want to pros pone this any longer and have some free time at the moment.

I should mention that I did this mod purely to examine any power/economy gains, not because I felt the factory GTI cooling system/water pump was insufficient.

Check out the website http://www.daviescraig.com.au for all the stats, marketing blog ect. I picked it up on EBay with the optional hand controller, well below retail price.

I had it installed around November 2006, so have been through various weather seasons with the setup.

Theory/Purpose

Running a shorter belt to the alternator which bypasses the water pump pulley frees up power and improves fuel economy.
Manual control of temperature can allow warmer settings for better fuel economy or colder settings for increased power.
A more effective water pump allows for increased cooling efficiency.

In performance terms I recall it being referred to as an alternative to say an underdrive pulley setup.

Installation

Due to time/tool access restraints and the fact my old pump was leaking I didn’t really want to stuff around. I had the install done by Suzitech and they did quite a good job of it, particularly considering it was the first time they have installed one.

My old leaking water pump was plugged with Chem Weld and its pulley is no longer utilised. A shorter belt runs from flywheel to alternator pulley.

The pump itself sits just above the passenger side CV Joint. It could probably be mounted anywhere practical within the cooling system but this is probably the best/only place on a swift. The manual suggests using support brackets if required but the 2 hoses provided and used are short and rigid enough to hold the reasonably weighted pump up by itself. Even pushing down on the unit I couldn’t make it contact the CV boot it overhangs, so clearance is satisfactory. Power runs direct from your battery.

The package included some adapters to fit the included hoses from the pump to the cars coolant tubes but these proved too big, so Suzitech had to weld some flanges onto the coolant tubes to accept the new rubber hose size. Obviously this increased the time/cost of installation.

The thermostat is done away with and in its place a thermo sensor attached to a wire (runs to pump) is placed within the upper radiator hose connection.

The temperature control unit was placed in the space underneath my centre console. It was secured with some double sided Velcro to the carpet. The unit is adjusted with a small flat head screw driver. Left is colder, right is hotter. The manual suggests the equated temperature ranges it can be adjusted within.

Overall I haven’t experienced any major leakage problems from the setup. I have found the odd spot but coolant levels have been consistent.

I don’t wish to go into figures but all up it apparently took around 8 hours of work. I was charged approximately 5.

Another option is to hook the pump up with your turbo timer so that when you shut your car down the coolant continues to pump for your desired time to prevent engine heat soak. Also possibly useful if your engine overheats and you want continued coolant flow to potentially avoid warping the head.

With the car keys on accessories alone the pump appeared to momentarily shutdown and thus not pump with the engine off but I’m not certain on this. The pump itself is quiet but it’s possible to make it out at idle if you listen.

Keep in mind that the electric water pump can be used in conjunction with your existing water pump/cooling setup (belt driven/mechanical) simply as a booster to cooling so hyperthetically could be connected/disconnected as you please. Ofcourse by doing this you are not reducing engine load, just providing extra cooling.

Results/Impressions

Cooling Performance

For the most part the electric water pump does its job of facilitating coolant flow, sometimes perhaps a little to well for its own good. When driving the car from a cold start the water temperature gauge takes longer to see some movement but when it does it rises fairly abruptly (compared to the gradually smooth rise on previous pump) and then sits at the half way mark without shifting just as it did with the factory setup and as it should.

However in the morning particularly when the car is coldest a rev hunting problem persists. When first starting up the auto choke will have the car idling high around 2000rpm as per normal but if you give it a minute whereupon the idle would normally start to fall the revs start jumping from 1500rpm to 2000 rpm. This continues as you drive away (obviously only in neutral or with clutch in). However the revs remain high even in gear. For example take your foot off in second gear and the revs at least in the immediate term will only fall to about 2000rpm and will thus continue to propel your car forward somewhat which isn’t fun when entering a round-a-bout in the morning.

My guess is that without a thermostat to restrict flow when cold, an abundance of coolant is pumped through the cooling system and the auto choke goes a bit crazy until the car reaches a decent temperature which does take longer than normal. It may be something else which occurred during or as a result of the install so I plan to get Suzitech to check it out but I don’t think this is the case, but a fix or solution may be available.

Such problems can at times be reduced/shortened in duration by adjusting the temperature control.

As she warms up the rev hunting gradually settles and isn’t as drastic until eventually depending upon the outside temp and the temperature settings you have set, the idle will settle down to a reasonably constant state but may still idle somewhat high, e.g. 1000 rpm. This is likely the auto choke compensating for the cooler water temperature. However despite this you will still hear the radiator thermo fan turn on/off as it sees fit.

In summer I have much less trouble, I just set the temperature control to the coldest setting and the weather does the rest. It probably still has a bit of a rev hunt when warming up from cold but I can’t recall entirely. Once warm though it tends to stay that way and it idles at or around 900rpm as per factory standards.

On some longer trips I have experienced the temperature gauge jump up high after which it would eventually drop back to normal before later repeating the spike. Adjusting the controller to a colder setting solved this problem.

I think it questions the setups ability to maintain a consistent temperature throughout various conditions of whether and driving.

The cars heater takes a significantly longer time to start working and even when it does it isn't very effective. In winter it's not warm at all but adjusting the controller for a warmer temperature mey help a small degree.

Car Performance

First impressions are that the car revs slightly harder in the upper mid to high range but I couldn’t really notice any real performance benefits first up.

However with some nice cool (i.e night) weather around and the temperature control set on the less warm and colder side of things I was happy to experience more mid range punch off the mark and bit more of a kick as she revved out in first. This put a smile on my face. Harder to really comment on the other gears in regards to top end power but my dyno results do suggest a little more peak power in the higher range was made.

It’s a bit like when your car is cold in the morning at has that extra power and response but on a more permanent basis and with more measurable and improved results.

On a hot day my cars power tends to feel the same (i.e. sluggish) as before.

Examining the 2 dyno sheets you can see that peak power has increased by 1.5 kws. More significantly examinig the midrange performance around the 130km/h mark you can see an increase of slightly over 3kw. So sure they arn't big numbers but most N/A swift mods don't add big hp figures to the table but every little bit adds up and helps.

So throughout the rev range we can examine at least a 4.5 kw increase in power. That's an approximate 6.5% increase in overall usuable power. Would be interesting to get it back down the 1/4 mile but couldn't predict it taking much if anything off.

It seems as though the car performs better with the aircon turned on as opposed to how it used to with the aircon on. Guess the engine has one less belt to strain it.

In regards to down low and up hill performance I couldn’t say I’ve noticed any improvemet which would suggest my tourque output (at least down low has not changed) . However it seems to go better with people in the car than it used to. But really my last 2 comments in particular really are just conjecture and seat of the pants, so I can’t really back them up. I also had a problem with my car occasionally stalling with the aircon on (revs falling to low). For some reason this problem has dissapeared.

In regards to engine sound the pitch of the engine when revving out towards redline doesn’t seem as harsh. It seems to have a sweeter sound. Not louder or anything I don’t think, just a different tone. This was a welcome little observation.

I haven’t noticed any change in fuel economy despite claims by the manufacturer that this setup can do such, however I haven’t run it at the hottest settings for any prolonged period either.


Perhaps on cars with larger engines the power/economy benefits are more profound.

Summary and further points

Pros

Pump itself is solid and seems well built. Australian made doesn’t hurt!

Has people point and ask “what the hell is that?â€

Fantastic write up there Gards

Have been thinking about doing this for a while and you have just saved me a wad of cash!

On the hunting, are you running an underdrive pulley? or an aftermarket chip? I had a problem with hunting which related to excessive power drain from accessories. Given you have added an electric pump, I thought that this could exacerbate this.

Original post is now updated with the 2nd Dyno Graph and extra discussion under car performance heading.

Thanks for the compliment grape, glad it's helped you with your decision and that others have been looking at this setup.

No, only other mod is unifilter, Chip will be goin in soon though. Wonder how it will react with that, probarly won't affect things.

You mention race cars, Ill stick with my belt driven pump. Less prone to failure than an electric motor.

For short races (1/4 mile, hillclimbs) i can see the benefit in not driving the water pump via the belt . On longer races/drives the alternator still has to drive it.

Direct drive has less losses in it along the way.

Your heater problem and revs up/down are both to do with removing the themostat,stick the thermostat back in ,as long as you have the std hose configuration still connected you wont have an issue,the only thing is you may have to reposition the davies craig temp sensor to a point were it reads the temp as the engine warms up,i.e. not in the top hose.

Your right in saying that as a mechanical pump is not as complex as an electrical one but I could probarly say our mechanical ones are possibly more prone to leaking, as us swift owners know!



Hey now there's an idea, remove the alternator belt and have no belt driven strain on the engine whatsoever! Ofcourse just applicable for motorsport as suggested or install one of those alternator cut outs at high rpm. Should be good for another 3-4 kw!



Thanks for the suggestion dave might look into that.

I recently had the EWP setup removed and got back the modified coolant pipes that were used for the setup. Below are some pics which show you the flanges that were welded on, sorry that some didn't come out that well.








Btw I did investigate placing the electric temperature sensor somewhere else in the coolant line as suggested by dave. However upon reading the manul I discoverd that the sensor goes way down towards the engine block in order to try and get as close as possible to engine temperature readings! Considering a standard thermostat and temperature gauge read of coolant temp, not engine temps no wonder things were strange, although that's probarly just part of the story. Due to this I didn't know how it would react by placing it somewhere presumbly cooler in the system away from the engine. I will reiterate that I didn't feel comfortable simply using a standard thermostat and running the EWP at full capacity on a permenent basis without the hand controller and electric temperature sensor.

Another option is to run the EWP in the standard cooling chain hooked up to a Davies Craig thermo switch set to a certain temp. This way the EWP will only kick in when a certain coolant temperature is reached such as 100 degrees in the event of overheating due to a failed standard pump or broken fan belt for instance. A handy backup that shouldn't interfere with your standard factory cooling unless it is really needed. But I deemed it not necessary for a street GTI with a reasonably reliable factory setup.

I also did a bit of reading and there is some theories that suggest an EWP will provide improved coolant flow at idle over a factory mechanical setup but at high revs a mechancial water pump will provide superiour flow.

For your information I contacted Davies Craig with a lengthy email detailing my problems months ago but to this day have not received a reply and I'm almost certain that email did actually get sent, lol. I have read some negative things about their marketing and support in my recent research.

So for ease of mind and to solve the problems I weant back to a factory setup. So thus ends the fabled EWP experiment... at least for now and on this swift as I have a feeling it may be popping up on someone elses in the future...

Thanks for reading!

We converted the rally car to one of these Davies Craig pumps. It is mounted in the bottom radiator hose, just near the gearbox. I have wired it up without all the extra accessaries, so it comes on with the kill switch and stays on all the time. We run a normal thermostat, so the electric pump is acting just like the mechanical pump.

I machined the water pump housing (removed the shaft and impeller) and welded it flat. We know run a much shorter belt, just to the alternator. This is where the HP gains occur, because you are not driving the pump.

Everything seems to be working fine like this since the begining of the year.

Glad to hear the setup is working for you. How did you go with hooking up the hoses to the coolant pipes?

A very informative & comprehensive write-up
Great work Gards!

I am researching a similar (but custom made) setup, and have the following queries:
a) What length belt is used for just running the alterator?
b) Can this not be dropped to a 3 rib in lieu of a 4 rib belt?
c) What diameter are the standard crank, water pump & alternator pulleys?

I think any street setup must retain the thermostat for driveability, and to enable practical use of the heater/demister.

Thank you for the support!

Unforntantly I cannot answer your questions. But if you ring Suzitech and talk to Pete who knows all about the installation/removal of the EWP setup I recently had he may recal what shortened belt they used. Just tell him you know Andrew Gardiner who had the EWP in his GTI.

I think it took him a little scrouging for him to find one that fitted properly.

From what I remember, we just cut a small length out of the bottom hose and used the different size fittings that came with kit.

great write up

i'm currently adding a electric waterpump from a vw transporter.... to give my watercooled turbo more flow, prevent it from boiling.... i'm curious how it works... the waterpump will be controlled by the adaptronic ecu, i'm thinking of triggering it on temperature that leaves the turbo, but maybe it's wiser to use it all the time because it's generating a lot of heat all the time... and of course to recirculate after shutting down... does anybody have experience with adding electric waterpump to our gti turbo? i tapped a hole on lowest point in de the radiator as the feeder with the pump inline, after the turbo it's connected to the old throttle body connection near the mechanical waterpump...

Had some people interested about the shorter belt that was used to bypass the water pump pulley. I.e. runs just from crank pulley to alternator.

Fortunatly as I was going through my EWP box I found that the belt was left over after the de-install.

It is a Gates Micro-V. Model number is K040275.

I was also reading the EWP manual and funnily enough did come across something that was suggested by a few people on redline and teamswift.

Under pump controll options it mentions running the EWP at continuous full capacity with the standard thermostat in place but it suggests a small 5mm hole in it which allows a small amount of flow to circulate while the thermostat is closed. So there you go.

Thanks to everyone for the interest of late, it has been encouraging.



There are relatively few people in general with experience with EWP's, yet alone ones with experience with swifts turbo or not. And it sounds as though yours is a different kettle of fish as it is a factory item and can be controlled by the ECU which I must say is impressive if it works that way. You are talking about a Swift Gti yes or is it a Golf Gti?

yesterday i finished the install, works great i'm talking about my turbocharged swift, i used a electric pump to increase watercooling flow thrue the turbo....

it's speedcontrolled by the standalone ecu from adaptronic on a 25hz pwm signal based on the temperature of the water from the return line.....

i can make some pics....

Heh nice work mate. Some pics would be interesting.