Electric Radiator Fan - to shroud or not to shroud?

[pkchopp]

Spal pusher fan with curved blades 

12” for roundies...

14” for square  

[Healey3000]

Hi,

 

Does a 14" fit flush with the radiator or is it offset to the front by a bit?

 

Thanks.

[Healey3000]

On ‎6‎/‎21‎/‎2018 at 3:09 PM, jimk said:

I think these fans have a ring on the blade assembly? that takes care of this.

Just put it in and run it.  Many are in pusher and no heating issues.  Besides the fan is only needed at speeds less than 40mph and at idle the engine is not putting out that much lost heat.

Mine is cooling a S14 using variable speed from a SSR relay and rarely runs at over 75% of full speed.  It reads the coolant outlet flow from the radiator, begins the ramp table at 158F and would be at full speed at 180F radiator out.

Hi Jim,

 

Do you have any pictures of your implementation of the temperature sensor?

 

Thanks.

[jimk]

2 hours ago, Healey3000 said:

Do you have any pictures of your implementation of the temperature sensor?

When I had the M10 in, I had a Silicon Garage radiator that I had him install a bung in the lower tank near the outlet.  With the S14 the sensor is located in the E30 thermostat.  There was a bung already there but unused with a plug.  Nothing that a picture would help with.  The sensor is a 2 wire, one for the ECU to read the resistance to ground, the other is the ground attached to the ECU signal ground that is shared with other sensors.

[Healey3000]

Thanks, that's a nice t-stat housing.  I wish the 2002 housing had room for a sensor.  I'll probably use one of those hose intercept adapter types.  I know I've seen one on the FAQ but I can't seem to find it now.

[pkchopp]

A early 3 searies water neck has plenty of room for multiple sensors 

[Healey3000]

3 hours ago, pkchopp said:

A early 3 searies water neck has plenty of room for multiple sensors 

True, but they are all on the output side of the head, not the return.

[jrhone]

2 hours ago, Healey3000 said:

True, but they are all on the output side of the head, not the return.

 

Doesnt matter if you get the right temp switch.  

[Simeon]

Aren’t they both ‘outputs’? One is used as the output to the rad (obviously) and the other provides hot coolant that is mixed by the thermostat and run back into the engine during warm up? I doubt there would be any functional difference in temperature between the two though noting actual flow through one would slow and stop as the thermostat opens. 

[Healey3000]

On ‎6‎/‎21‎/‎2018 at 6:07 PM, jimk said:

Way too low.  I didn't know what to set it at and have been running at a very low 100 Hz.  Info you found is the only frequency I have seen.  So I took a look and the max I can set it to is 12kHz.  So I'll try it there to see if it stops growling the very low duty cycle (or if it runs at all??).

Jim,

 

What brand/model of SSR are you using?  Most, if not all SSR's that are designed as high-side switches have very poor frequency response.  If so, yours is probably having a devil of a time running at 12 kHz and may be getting very hot, with a much lower voltage reaching the fan.  I grabbed a Chrysler minivan SSR from the junkyard and it is unhappy any higher than 100 Hz.

 

Looking at available MOSFETs and high-side automotive devices, all the P-channel stuff is slow.  There are very fast N-channel MOSFETs that would manage 12 kHz without a problem.

[jimk]

1 hour ago, Healey3000 said:

What brand/model of SSR are you using?

I have it mounted against the inner fender flat area with heat conductive paste.  Where there isn't a way to transfer heat away, there is a heat sink sold for it.  When I bought mine, heat sinks were included, but it's stored.  I can't say it is or is not getting hot since I upped the frequency, but I drove today in traffic across town to get exhaust work.  It runs quiet now at low duty output.  Wish me luck on the potential relay overheating.  Am using a 40amp, and have two 10s and a 25 on the shelf that I was using for another purpose on another car.

https://www.ebay.com/itm/US-Stock-40A-Solid-State-Relay-SSR-DC-DC-Input-3-32V-DC-Load-5-220V-DC-DD220D40/391713162983?hash=item5b33ec86e7%3Ag%3A6yAAAOSwhGlZmjX1&_nkw=ssr+relay+dc-dc&_sacat=0&_from=R40&rt=nc

 

FWIW Haltech says to wire it this way and so I did

 

Solid-State-Relay-HT030202.pdf

Edited July 26, 2018 by jimk

[Healey3000]

24 minutes ago, jimk said:

I have it mounted against the inner fender flat area with heat conductive paste.  Where there isn't a way to transfer heat away, there is a heat sink sold for it.  When I bought mine, heat sinks were included, but it's stored.  I can't say it is or is not getting hot since I upped the frequency, but I drove today in traffic across town to get exhaust work.  It runs quiet now at low duty output.  Wish me luck on the potential relay overheating.  Am using a 40amp, and have two 10s and a 25 on the shelf that I was using for another purpose on another car.

https://www.ebay.com/itm/US-Stock-40A-Solid-State-Relay-SSR-DC-DC-Input-3-32V-DC-Load-5-220V-DC-DD220D40/391713162983?hash=item5b33ec86e7%3Ag%3A6yAAAOSwhGlZmjX1&_nkw=ssr+relay+dc-dc&_sacat=0&_from=R40&rt=nc

 

FWIW Haltech says to wire it this way and so I did

 

Solid-State-Relay-HT030202.pdf

Hi Jim,

 

Based on the link to the one on ebay, these are slow relays (<10 msec, therefore 100 Hz).  Haltech also says to set DPO output to 100 Hz. If you have an o'scope you may want to take a look at the output waveform to confirm. In my opinion, I think you'd be better off living with the noise at 100 Hz than risking relay failure and possible engine damage. 

 

I'm prototyping a fan PWM circuit and plan on using a fast N-channel MOSFET that will handle the high frequency that these fans seem to like.  I'll keep you posted on progress.

[jimk]

45 minutes ago, Healey3000 said:

I think you'd be better off living with the noise at 100 Hz than risking relay failure and possible engine damage.

Your advice is good. My memory is short, but Haltech's dwg may be the reason I had it set to 100 Hz until this subject came up on the FAQ.

I was doing some testing in the last 20 minutes and discovered there is no fan controllability above a commanded 10% duty cycle at 12000 Hz.  The fan would go to full speed.

So as I reduced the frequency, the controllability began to get better as the frequency.  But even at 700 Hz when the commanded duty was greater than about 40% the fan would jump to 100% speed.

At 150 Hz there is variable speed up to about 80% commanded duty where it goes full speed.

I'll back off some more on the frequency to 100 Hz and just explain the noise is because the car wants to go full throttle!

Thanks for the advice.  It's good to have your electrical knowledge available.

Jim

Edited July 26, 2018 by jimk

[TobyB]

On 6/27/2018 at 9:18 AM, jrhone said:

 

Doesnt matter if you get the right temp switch.  

Does matter, as ideally you'd like the temperature in the outlet

to be stable, while only turning on the fan when the radiator was 

starting to heat soak.  Thus, a rising engine INLET temp is the indication

that the fan needs to come on.  A rising OUTLET or water jacket temp means

that your engine is overheating.

 

Matter that much?  Obviously not, since it works,

but it's harder on things like head gaskets.

 

t

mechanical fan

[Vicleonardo1]

Need a bit of clarification. Why do the Spal controllers suck? Any reason to replace one that is working?