Cooling fans for BakerBM/Radtec style rad

[jackherer 543]

My 205 has a BakerBM/Radtec style low profile double thickness radiator.

 

I've never been entirely happy with the cooling fan on it, generally it is OK but when pushed to the limit it can't cope.

 

Last time I had a problem with it I fitted a second fan but it's still not enough.

 

When I say pushed to the limit I mean e.g. driving along a motorway and then suddenly stopping due to a traffic jam followed by a lot of start/stop edging along. The temperature gradually creeps up and turning the cabin fan on slows it down but it still creeps up. If I am then able to go over 15/20 mph it cools down immediately.

 

It's fine in all other circumstances so I'm sure the cooling system is working well, it's just the fans that can't quite cope.

 

The fans I have currently are the cheap generic 9" type. I'm thinking now that they aren't powerful enough to push air through the double thickness rad core...?

 

Is it worth spending a load more on a SPAL or something else branded? I don't want to spend £120 and find it is no different to a £10 fan.

[Tom Fenton 1,541]

I found the same and fitted a spal on the engine side. Yet to really try it in anger but it seems to do the trick.

[jackherer 543]

That's reassuring to know, I'll spend some money then!

 

Any reason for fitting it engine side? Is it more efficient or something?

 

I've got them in front blowing currently.

[Andy 105 1 Cars]

I did the same thing and fitted a second cheapo fan from MgGill motor sports for about a tenner . Had the same issue that they worked really well 95% of the time , but did struggle in the odd motorway situation . However, never enough of one to,

prompt me into doing anything about it!

 

[Andy 105 1 Cars]

Pulling is more efficient. , plus the fact that the  fan and its casing is impeding the airflow after the air has passed

through the core rather than before . Snag with pull is that clearance is a bit more of a problem . Mind you, Peugeot managed with a push, although the the o.e fan is pretty big fan with a motor than can draw some serious current 

[Tom Fenton 1,541]

Supposedly more efficient that way round, I read somewhere that with the fan in front of the rad, that area can only ever flow the fan max flow.

Also because the spal fan i bought second hand was a pull type so needed to go on the back.

The difference in flow is massive, look at the cheap ones they have a huge boss in the middle of the fan, this blocks airflow.

[jackherer 543]

7 minutes ago, Andy said:

I did the same thing and fitted a second cheapo fan from MgGill motor sports for about a tenner . Had the same issue that they worked really well 95% of the time , but did struggle in the odd motorway situation . However, never enough of one to,

prompt me into doing anything about it!

 

 

it's nice to know I'm not doing something wrong then!

 

As you say it isn't a common situation, it's happened maybe two or three times in four+ years and I've changed something each time in the hope of solving it.

 

However I bottled it last night in a traffic jam and shut it off. The problem is my race battery will only start it once or twice so I can't keep turning it on and off and that's not good for it anyway due to hot spots forming. I got out to warn the coach driver stuck behind me that it might take me a few seconds to get it going when the traffic cleared but he was more interested in asking if it was a 1.6 or 1.9 and telling me he used to have one...

 

OK I've noted both of your points re: pulling rather than pushing, I'll get a SPAL ordered ASAP.

[Tom Fenton 1,541]

I had to lean the rad forwards a bit to clear the throttle body on my Mi16 car. I welded some brackets on slightly further back.

 

Just as a point of interest, I did try a yellow band single speed fan switch. I think this must be off a diesel. Cuts in nice and early but the fan wasn’t enough to get it to cut out again. As it’s wired permanent live I was also concerned about it flattening the battery. I’ve gone back now to a red band twin speed switch but as per bloody usual the slow speed has bloody packed in, so it gets further up the gauge than I’d like. I’m so fed up with these bastard switches, I must have fitted half a bloody dozen in the last couple of years, just to my cars.

[jackherer 543]

It's crazy the switches are that bad now, I'm sure they were OK back in the day on the whole.

 

Mine is triggered by the Emerald so I've dodged that issue.

[jackherer 543]

I got a SPAL fan delivered a few days ago and as you say it is obviously a lot more powerful.

 

In the end I got another blower and I'm going to fit it in front.

 

That was only because I found a blower in the right size for a great price.

 

https://www.ebay.co.uk/itm/113720129661

 

I haven't permanently fitted it yet but the dimensions are identical to the cheap noname one I am replacing (except it is at least twice as deep) so it will fit using the same mounting points as the old one. The clearance to the bumper is minimal but it does fit.

[jackherer 543]

This fan was fully tested last night, some total idiots broke down in the slow lane of an unlit dual carriageway and their mates stopped in front of them to jump start the first car. Both cars had no lights on and a massive traffic jam had formed behind them that I got stuck in.

 

The new fan was perfect, it was cutting in then cutting out after only a few seconds because the temp was dropping so fast, I might even have to change the settings in the ECU to compensate. Admittedly it was late at night and ambient temps were much lower than last time this happened but the difference is clear, the slimline weak fans are not up to cooling the double thickness rad and the new one is.

 

I also discovered that the temps on my SPA dash are 2-3 degrees higher than my ECU sees so it might not have been quite as hot as I thought when the old fan wasn't coping.

[Andy 105 1 Cars]

Glad all is working well. As for ecu vs digital dash differences in temp reading, I had quite a job calibrating my ecu( Bosch) temp sensor and in the end made sure that both agreed with each other at various  set points, admittedly working on the assumption that the Stack sensor was accurately calibrated and matched to the reading . The problem with the Bosch sensor is that by 90 degrees , the resistance is between 180 and 200 ohms and forms part of a voltage divider , the other arm of which is internally set in the ecu at 2.7k . That means that at 90degrees, the voltage reading for the ecu is below 0.5v ( 5v stabilised supply)  and therefore very sensitive to very  small pd fluctuations . The ECU should have used a 1k internally . 

Ramble over. Glad your system works. 

[jackherer 543]

That's interesting. I would naturally assume a Bosch OEM sensor to be superior to an aftermarket 'motorsport' part but I didn't consider the ECU itself as a factor!

 

I just had another look and it's actually a bigger discrepancy, more like 6 or 7 degrees. I also have a Racetech mechanical gauge and that's about half way between the two...

 

I might do some testing to find out which is closest to reality.

 

Next weekend I plan on driving it to Nottingham to see Dr Sarty's Xsara so having the fan fixed is a relief.

[Andy 105 1 Cars]

Ah. The doc is in Nottingham now? Please give him my best wishes. I trust that if you are off to see him, his Xsara must have made some progress. As for the Bosch sensor , my first set of experiments was a mug, multimeter , thermometer and boiling water . I plotted a cooling curve. The snag is that the resistance range from even 20 degrees to 100 degrees is about 2.5k , Worse is that by 90degrees , the resistance is around 200ohms and  very sensitive to small temp changes . That would not be a problem were it not for my ECU designers choice of internal resistance. On the Emerald , the internal resistance is 1.0k , which is a much more sensible choice for this type of sensor , as at 90 degrees , 200/1200 is around  0.82v . For mine, 200/2900  which gives 0.35v ( roughly)  0.05 v variation or a couple of ohms on the resistor and the temp reading  alters significantly . In my view, Emerald made a good choice . My way out would have been to use a Zetec WTS that has a resistance range of 12k down to 1k . Anyway, what I have works , so will leave well alone ,but iit was a few hours of experiment to get it all to work reliably. Oh, I ended up using screened cable for all sensors to reduce any spurious voltage signals . 

[Andy 105 1 Cars]

Almost  forgot. The second part of the sensor calibration story included a trip to the Bosch site to get the equation that describes the resistance - temperature relationship for the thermistor. Exponential expression with two constants , one in the exponent . Readings of resistance at two different temps, quick play with the pair of equations to find the constants, and then use the equation to ‘predict ‘ the resistance temp curve. Have to say that it worked extremely well and was a great way to correlate against the experimental readings . 

[jackherer 543]

Ah OK, for some reason I was thinking you had an Emerald but now I recall your issue with the firmware for the cam sensor not being right, which ECU is it you have?

 

I can see I have some research to do now, my coolant probably needs changing anyway so I'll take the sensors out and see what I find. Where did you see the equation you're talking about on the Bosch site? All I can find are datasheets with tables of resistance vs temps. BTW I noticed that the accuracy at 100 degrees is +/- 3.4 degrees.

 

Yes, Rich has got some stuff done to the Xsara, and the plan is to get stuck into it this weekend so hopefully he'll post a substantial update afterwards.

[Andy 105 1 Cars]

Hi. The ecu is a ME442 from Motorsport electronics . For the equation, I will look it up and either send the link or the actual equation . Rich has made some progress then. It will be an interesting and exciting project .

Andy

[Andy 105 1 Cars]

The equation is as follows,

 Rt = R25xEXP(b(1/(t+273)  -1/(25 + 273)) 

where Rt is the resistance at temp t degrees  and R25 is the resistance at 25 degrees . b is a constant which is a function of the type of NTC thermistor , but can be determined as I suggested above .

Andy

p.s 273 should really be 273.15 , but......