[MaxRothery]

Hi All

Im very new to the Speeduino forums and have a slight understanding of wiring but nothing on coding.

Im doing an engine swap and have been looking at using the BMW electric water pump for simplicity of the ancillaries belt on the engine.

The pump uses PWM for speed control, and my understanding is that the current firmware only allows for the cooling fan to be on/off operated. But I noticed the idle air valve uses PWM for control, which I’m not using, so in theory, could I use the idle air control pwm as the controller for the water pump or would I have to use an aux output to an external controller to power the pump?

I know this wouldn’t be useful for a lot of people as they’d probably use the idle air controller as intended.

For those interested, the engine swap is a Jaguar S Type 3.0 V6 with home brewed ITBs into a Triumph Dolomite. The ITBs are why I’m not running an idle control valve.

[dazq]

Hi, looking at the data for the pwm it may work.
https://www.tecomotive.com/download/PWMinfo_EN.pdf

I am planning to use one of those on my Jag aj30 also as the s type pump protrudes far too far forward.

Keep us posted how you experiment goes

[PSIG]

Interesting control requirements. Simple but specific, and easy to botch if you don't know. Thanks for posting that, @dazq!

[punisher454]

PWM speed controls with an arduino are super simple. I've been using variable speed volvo fans on several projects and they are basically the same thing, just a pump instead of a fan. Same basic pwm concept though.
I'm in the middle of building a project right now that uses a Pierburg CWA400 for a 350 chevy in a v8 S-10, I'll report back on this forum about the results probably later this fall.
I'd personally recommend a standalone controller right now rather than trying to add more stuff to speeduino. I know some people dont use thermostats when they run electric pumps, and some use restrictors with the idea that the restriction will help create a higher relative pressure in the engine with a pressure drop going into the radiator. Personally I'm considering using a 160degree high flow thermostat (with a small 1/16 to 1/8 inch bypass hole drilled. my actual target temp will be like 190f for cruising, and down to 160f when I'm looking for power. I think you should always maintain a minimum pump speed, and then at whatever your desired running temp vary the pump speed to maintain it. No need to add this overhead to the speeduino. Probably a wise idea to add a manual switch for full pump speed if you have a sensor failure or other emergency cooling problem.

[PSIG]

… I'd personally recommend a standalone controller right now rather than trying to add more stuff to speeduino. … I think you should always maintain a minimum pump speed, and then at whatever your desired running temp vary the pump speed to maintain it. No need to add this overhead to the speeduino. Probably a wise idea to add a manual switch for full pump speed if you have a sensor failure or other emergency cooling problem.

Ah, simple, yet there are added requirements. I completely agree with standalone as it allows freedom for any functions you prefer, changeable on your demand, while also not potentially compromising or limiting engine operational code. Win-win-win.

In this case, the Pierburg CWA pump needs more than just a PWM signal, including a wake-up signal, and automatic full-speed override, but only outside of DC limits, which also ensures minimum pump speed with fail-safes. All good stuff , and not complicated, but deeper than just a variable DC Arduino PWM signal based on temperature (serial, CAN, or direct). Users post your findings when you get there!

[MaxRothery]

I completely apologise for not getting back sooner, for some reason, I haven’t gotten any notification emails for this.

For various reason, I have held back on this as I think I was playing around in tunerstudio and it came up with the error that the max pwm frequency was 512hz and the input frequencies for the pump was 1000hz, meaning I wouldn’t get the full range of control. This could just be my bad memory playing up on me.

I did think about having a stand-alone controller, as peirburg sell their own, but I’m a tight bastard and like to try and outthink a problem rather than pay it away.

Hi, looking at the data for the pwm it may work.
https://www.tecomotive.com/download/PWMinfo_EN.pdf

I am planning to use one of those on my Jag aj30 also as the s type pump protrudes far too far forward.

Keep us posted how you experiment goes

Yea, I’d does stick out an awful lot. What you swapping the engine into?

[dazq]

The engine is going in a one off hot rod SUV, along with a couple of turbochargers.. ..Yum

[MaxRothery]

The engine is going in a one off hot rod SUV, along with a couple of turbochargers.. ..Yum

sounds like a fun project!! a hot ford SUV sounds like a challenge. V6 should take to boost well

[PSIG]

… the max pwm frequency was 512hz and the input frequencies for the pump was 1000hz, meaning I wouldn’t get the full range of control.

You might be assuming some stuff? Per the Pierburg CWA pump datasheet, the duty cycle (DC) appears to be the control mechanism. It is possible the 50-1000Hz spec is simply that it will accept any signal frequency in that range.

To show the point, the datasheet example of a full-on wakeup signal is 150Hz (easy way to get >3ms "on" signal without additional timer), not that it must be 150Hz. I would test the pump at various values of Hz and DC for verification, as it would greatly simplify your job. Good luck!

[punisher454]

I've got a couple projects in the works that both will be using the CWA400, So Ive been watching this.
What I have done in the past when using an arduino (or just a bare atmega uC)to control an OEM thing like a volvo variable speed fan, is to set up a pwm output pin with the time configures so I can adjust both duty cycle AND frequency. Then I just start by testing the device at various frequencies with the duty cycle around 50%. I make a note of what frequency range the device likes, then I test to both the upper and lower limit of the dutycycle range.
For example the bosch controller on the Volvo fans has a dutycycle range of something like 35-93% (dont remember EXACTLY what it is) and I think I recall it worked fine at 100hz.
The Pierburg is very likely similar to the bosch fans in that the PWM is the speed control and the frequency just needs to be in the right range to accept the signal.
Several people on youtube have built simple arduio based controllers for the pierburg pumps.
just watched a guys youtube video of controlling one with rusefi. He was using 500hz and runing up to 85% pwm, 8%pwm was stopped and it looked like his table had the pump strting at 14%.
https://www.youtube.com/watch?v=4Ee9d1lMbzQ
Another video seems to show the pump working at max speed at 97% dutycycle.

This video shows a cwa400 being controller by a 150hz signal, up to 100% DC (so maybe you can control it with a 100% dutycycle)
https://www.youtube.com/watch?v=Y7hpfvztm0M

This guy shows the CWA400 starting at 13% dutycycle , and i think it was at 12V on the pwm (not sure might have to watch again).

OK just found THIS video
https://www.youtube.com/watch?v=atLzbKPMnnU
This guy shows how he sets up one on a terminatorX. You should be able to do the exact same thing in the speeduino.
He also compares it to the meziere and shows his instalation on a 6.0l turbo LS engine in his truck that he drags. VERY good stuff!!!