Amps, Alterntors & Rad Fans

[ChimpOnGas]

So my N/S rad fan is sick, on start up it's amp draw is spiking considerably above what's acceptable the result being it occasionally welds the relay contacts together resulting in the fan running continuously.

Even before the relay fails I can see a significant drop in volts as the suspect fan chimes in which clearly has an affect on injector opening times as I can see my AFR go lean at the very same moment the fan fires up, I solved this with the voltage correction feature in my Canems software designed for this very issue but whole experience got me thinking

Clearly the sickly fan issue needs addressing, replacing it with a new but same spec fan will solve the problem but as I like to take such failures as an opportunity to upgrade I plan to replace my troublesome fan with one of Spal's newer brushless motor type which also offers soft start technology. These brushless motor fans draw less amps overall but also eliminate the typical big amp spike you can see with traditional rad fans when they first kick in.

All this got me looking at how many amps the car draws at idle, so assuming I'm not using my LED Headlights or LED driving lights I thought it would be interesting look at the true electrical load my 100amp alternator needs to support at idle keeping in mind an 100amp alternator will only serve up 50amps at idle at best.



My figures seem to suggest without turning my headlights or driving lights on my 100amp alternator has things covered, even with all the lights on and running LPG it should just cope, but as soon as you throw the rad fans into the mix the story becomes very different.

Upgrading from the standard 100amp alternator to a later Gems P38 120amp (AMR2938) alternator would help improve things by 10amps at idle and a brushless fan reduces the draw by 4amps each so if I fit two of them that's 8amps less plus the 10amp from the 120amp alternator means I can claw back a 18amps. Clearly this is a useful improvement and from my calculations while it still doesn't quite completely solve the deficit, the two mods together would certainly help.

The evidence seems to suggest fitting two brushless fans and the 120amp (AMR2938) alternator is a good way to improve the situation so I think that's what I'll end up doing, I may still need to add a small mount of fuel using the voltage correction feature in my Canems software but I guess that's exactly what it's designed for.

I appreciate I'm only looking at idle conditions here, and even then the amp deficit is only present when the rad fans are functioning, at all other times even the standard 100amp alternator is proved to be sufficient. But at idle with both fans running it seems we could all benefit from a set of these lower amp brushless rad fans and a 120amp alternator.

Discuss....

Edited by ChimpOnGas on Sunday 27th November 12:00

[ChimpOnGas]

Discopotatoes said:

What is the amp draw when it spikes? Do you have two stage fan set up so the start up spikes are spread? I only ask as I've got 3 fans all set to come on at the same time and i was having alternator problems

Yes the fans are staged Rich, the otter switch runs one and the ECU the other, I never actually measured the spike but it's eaten three relays in the last year so there's definitely an issue.

Today I pulled the frozen relay and ran the car on the one good fan only, turns out in this colder weather the one fan system works just fine, coolant temp tops out at 90 and while sat at sustained idle the one fan can't quite get things down to it's off temp it does bring things down to 88 degrees and holds it there just fine. As soon as the car is moving over 15mph there's enough air flow through the rad to lower the coolant temp sufficiently for the fan to switch off.

With the engine off battery voltage is 12.7v, so all good but at 950rpm idle battery voltage only rose to 13.8v which seemed a bit short to me even with my low resistance Odyssey AGM battery. So I jacked the idle to 2,000rpm and battery voltage remained at 13.8v, I'd expect to see at least 14.2v so I waited for the one fan to cut in to wake the alternator up but all that happened was battery voltage fell from 13.8v to 13.6v.

I conclude I've either got a wiring issue or my alternator is on its way out, as I've already replaced the main battery lead with something better than TVR gave us and added an additional earth from the engine block directly back to the battery my suspicions are falling on the alternator.

Even with a low resistance Odyssey AGM battery 13.8v at idle seems low, but when the fan cut in I certainly would have expected to see 14.2v as a minimum, seeing the meter fall to 13.6v with the one fan running definitely seems to point to a weak alternator to me?

What do others think of my figures?

New alternator needed?

Edited by ChimpOnGas on Sunday 27th November 18:55

[ChimpOnGas]

Discopotatoes said:

were you taking the voltage from the battery or the alternator?

I see where you're going with that Rich, and yes I took my readings direct at the battery using my multimeter.

The multimeter readings match exactly what my Canems software shows, I also ran a log and no matter what rpm (idle to 5,100rpm) and over many miles of mixed driving my voltages never once went beyond an absolute maximum of 14v, the ECU is of course also taking it's voltage reading direct from the battery.

Thoughts

Edited by ChimpOnGas on Sunday 27th November 19:16

[ChimpOnGas]

Discopotatoes said:

then you know what I'm going to say next
take reading from the alternator and earth it to the body of the alternator, if you get your 14.5 volts or there abouts id suspect a resistance somewhere else, like your starter motor

Exactly mate

It got cold and I had beer waiting for me in the fridge so alternator testing got left.

Do you agree my figures are low?

I'll do my test at the alternator tomorrow, if I get the same readings direct at the alternator perhaps I've lost a diode, if I see 14.2v or more at the alternator I guess we have to assume a cable issue.

[ChimpOnGas]

Discopotatoes said:

i had the exact same thing on my p38. 14.5 at the alternator 13.8 at the battery, the only thing between them was the starter, i swapped alternators and got the same result.
id also be testing either side of the 100 amp fuse

Thanks mate, I'll do the checks, while 13.8v is minimum charging voltage it certainly seems to be enough to keep my battery charged, I also have no starting issues whatsoever, in fact the engine starts very energetically indeed.

So with no starting or charging issues why am I even bothered? well what concerns me is when the rad fan comes on I see a voltage drop from 13.8v to 13.6v at idle, TBH as the fan fires up I would expect the alternator to kick in and I'd see at least 14.2v perhaps even 14.5v... not a drop!

Directly linked to this drop in volts is a condition at idle where the AFR goes lean, I've corrected this using the Canems voltage correction feature where I'm able to add some fuel to richen the AFR back to my 13.7:1 idle target. This works well but I can't help feeling I'm masking an issue rather than fixing it?

Clearly more testing needed

[ChimpOnGas]

Belle427 said:

When I upgraded my battery to starter cable I noticed the alternator cable also looked inadequate in size and a little crusty. Have not got round to changing it yet.

When I presented my starter cable upgrade way back in 2012 I also looked at the alternator cable, I felt it was fine so left it alone, but perhaps that was a mistake?

http://www.pistonheads.com/gassing/topic.asp?t=110...

I replaced my starter motor, the long thin and rather crispy original starter cable with something much better and of a heavier gauge, and finally the short run from the starter to the 100amp fuse which I replaced with a physically much stronger Midi fuse.

Other than an additional earth from the engine to the battery I must admit I stopped there, the whole lot made a noticeable improvement and the starter motor span way faster, I felt the exercise was a great success but perhaps I should have pressed on and replaced that last run to the alternator?

Maybe linking my engine block earth to one of the the alternator mounting bolts would help too, like this I'll have a direct cable link from alternator to battery earth terminal. With that last run of cable from my 100amp midi fuse to the alternator uprated with something of heavier gauge too hopefully I'll finally see the 14.2v at my battery and ECU?

Edited by ChimpOnGas on Sunday 27th November 21:13

[ChimpOnGas]

Discopotatoes said:

do you have a link to the brushless fans Dave?

Yes mate, here's ProAlloy's eBay listing, they seem to be the cheapest source but maybe even cheaper if we arranged a group buy away from eBay?

http://www.ebay.co.uk/itm/322316912826?_trksid=p20...

The brushless motor isn't actually super new tech, Spal switched to it over 10 years ago, that's still a long time after our cars were made though.

My research shows each Spal 16" brushless motor fan will pull anything from 4 to 8amps less than the older brushed Spal fans I believe were used by TVR, and this is just during normal operation. They do shift a bit more air too but that's really only because brushless motors can spin faster, the big advantage for me is they also have a "soft start" feature. So even if your ECU doesn't support running a fan speed ramping effect using PWM you can still enjoy the considerable benefits of no volt/amp spike on fan start up.

As a minimum I want to see 14.2v at my battery & ECU before I go buying fancy fans, but saying that the regular relay failures I'm suffering do seem to suggest I need at least one new fan? The Spal 16" brushless motor fans are definitely better by design and certainly offer some worthwhile advantages so unfortunately I can see some more spending on the horizon

I've discovered ECU's are sensitive to voltage changes and especially don't like low voltages, this has a knock on effect to injector opening times which in turn can change AFRs and lead the engine going lean & hunting slightly at idle. This is clearly why you get the voltage correction feature with all after market ECUs and I'd be surprised if even the 14CUX wasnt doing it too, it certainly works on my Canems system.

But surely a much better solution would be for the ECU to see consistent 14.2 - 14.5v battery voltage at idle as you would expect and I've always recorded on all the other cars I've tested with a healthy alternator.

We all know TVR wiring is a bit iffy to say the least, but surely 14.2 - 14.5v battery voltage at idle is not too much to ask is it?

Edited by ChimpOnGas on Sunday 27th November 21:50

[ChimpOnGas]

Sardonicus said:

Have you checked the continuity of the ground/earth on that offending fan Dave? I dont need to remind you of the effect on those contacts if its bad, sure you have covered this base just throwing this out

Cheers Simon, no is the answer so good call mate, actually I just pulled the relay with its welded closed contacts, I'm currently running just the one fan.

With the one fan I'm still only seeing 13.8v at idle which drops to 13.6v when it cuts in, I raised the revs to 2,000rpm but the the battery refused to show any more than 13.8v with the remaining fan on or off?

My logs show the ECU only ever sees a maximum of 14v no matter what the revs are doing, these figures seem a bit low to me, I would have thought I'd get 14.3-14.5v at the battery?

Do you agree?

[ChimpOnGas]

Thanks Steve, very helpful indeed

[ChimpOnGas]

carsy said:

Dave, Bluebottle wrote this on an earlier thread. Not sure how they compare price wise to where you are looking at.



If you are looking to replace go to Spal direct far far cheaper than all the usual suspects.

Tel:[phone]

Ive just replaced mine with the most pwerful ones that will fit in the Griff nose (its limited depth due to the radiator angle and bottom chassis rail.
The ones i bought are VA10-AP70/LL-47A and i paid £71.46 (plus VAT) each for them

That's very useful Ian, thanks

[ChimpOnGas]

450Nick said:

I decided to rip out all of the wiring in my car and replaced it with the wiring diagram I drew above. I used a similar fan switch, then used a 70A DPDT relay from Farnell for the changeover and a single 70A relay for initial switch on. Works a treat!

http://www.ebay.co.uk/itm/Radiator-Fan-Switch-Visa... (fan switch)

Edited by 450Nick on Tuesday 29th November 09:49

Thanks Nick

Let me see if I've understood this correctly

Wired as shown your fans are in series for the first stage and in parallel for the second? Both fans initially come on gently at half speed which should significantly limit current draw on start up and mean they are quieter too, when they shift to full speed the current spike is also reduced because the fans are already running?

To make all this work I need a 70A DPDT changeover relay and a twin value otter switch?

I really like your setup for the following reasons:

1. It uses a proven, reliable and cheap to buy twin value otter switch

2. I get to free up my Canems ECU output that's currently controlling one fan so I can use it for something else

3. The fans soft start reducing the current spike we see on start up with the standard setup

4. The first speed is much quieter and should be sufficient to control coolant temps in most situations on its own without needing to resort full speed

5. In summer stop start traffic when full speed is required you still enjoy a greatly reduced current spike because your fans are already running at half speed

That adds up to five clear benefits, if I've understood your system correctly wiring the fans this way looks to me like the definitive setup

Thanks again, Dave.

[ChimpOnGas]

stevesprint said:

Nick
Thanks, that's a brilliant & clear diagram.

Here’s a few M22 x 1.5mm sensors to choose from with some old Quinton Hazel and Intermotor part numbers.

QH Part   1stStage  2nd Stage                       Equiv  
XEFS202    95-90     100-95    4 Pin               FAE37960   50018   Escort 1.6/8 90-95 Mondeo 1.8D 
XEFS40     93-88      97-92    3 Std Spades        Intermotor 50102 205 1.6/9 83-90, pug/Cit 86-93 
XEFS86     92-87      97-92    3 Pin Junior Timer  Intermotor 50092 205 1.6/9 90-94, pug/Cit 87-96 
XEFS31     84-79      88-83    
           87-75      93-82    3 Pin VW plug       Intermotor 50025
LFS025     88-83      92-87    3 Std Spades        Intermotor 50091 or 50216

I like the idea of this one with the modern 3 way junior timer plug plus its temperatures are about right for the 14CUX, any cooler and the lambda probes won’t work correctly.


Dave
I like your summary, you could even add "6. Halves the wear as the fans run at half the speed".

What temperatures does your Canems ECU turn your fans on and off? For the last 20 years I’ve always used fan switch 92-87 97-92 and found the first stage perfect for most of the year, the second stage very rarely kicks in like on the hottest day of the year or on a rolling road.

To answer your question I've configured my Canems system to turn on one fan at 89c and off at 87c, this is a very tight 2c on/off window because the otter switch controlled fan does most of the work and only needs the support of the Canems controlled fan very briefly.

I believe the rating I chose for my otter switch was 88c on and 83c off so the otter switch controlled fan always comes on first, the Canems controlled fan typically chimes in a few seconds later as the coolant hits 89c, it only stays on briefly while the other fan continues the work of getting the coolant below 83 degrees before that too shuts off.

With an 88 degree thermostat the car maintains coolant temps between 82 & 85 degrees while on the move depending on ambient temps, this means the fans are only ever used (and needed) when the car is at idle. The fans cycle on and off in a staggered way and never fire up at the exact same moment, this reduces the current drop associated with two fans starting in unison. Saying that one of my fans is sick so I'll be replacing it very soon with a new low amp brushless motor model from Spal with the soft start feature, clearly this will improve my system still further.

The otter switch controlled fan is on a permanent live while the Canems controlled one is running from a switched live just as TVR wired the original set up. This means when I switch the engine off the Canems fan will be stopped immediately but the otter switch controlled fan will continue to run until the coolant temperature falls to 83c, the permanent live circuit also runs a Davies Craig electric booster water pump so all the time that fan is running there's also a circulation of coolant around the engine even though it's been completely switched off and the mechanical water pump is no longer turning.

The system works well as coolant temps never exceed and absolute and momentary peak of 92c at idle as tested on the very hottest summer day this year in the South of France where ambient temps soured close to 100, under normal weather conditions the car will idle happily all day never exceeding an absolute maximum of 88c. Running temps also never exceed 85c with road speeds above a modest 15mph, and finally even with the engine completely switched off on a hot day there's still both air flow from one fan and water circulation from the electric booster pump, this system eliminates any risk of heat surge to the heads on switch off.

I still like the way Nick wired his fans though, and may well incorporate it as an enhancement to my current set up, but first I need to replace the sick fan that's causing its relay contacts to stick closed.

Edited by ChimpOnGas on Tuesday 6th December 19:05

[ChimpOnGas]

stevesprint said:

Dave
Thanks for your detailed answer, I’m impressed with your attention to detail and fine tuned Canems fan output, you’re nearly as anal meticulous as me

The only advantage of a Chimmy compared to a Precat Griff is the Chimmy rad is upright where as the Precat Griff rad leans forward more than 45 deg and therefore requires more speed to keep the fans off.

My fans are wired the same as 450Nick's and I wouldn’t have them any other way, its the oldest trick in the book for twin motors, I've seen the same wiring trick on a kids battery powered car with two gears, it works a treat.

Your & 450Nick's system is very elegant, mine is more simple than I made it sound as its really just an otter switch controlling one fan more or less as normal and my ECU controlling the other which makes it easy to tune.

The slow speed of your & Nick's set up is something I don't have and is a really worthwhile addition, I'm keeping it simple with mine by just adding one of the newer soft start Spal fans.

Both system are proven to work well it seems but your & Nicks is the better of the two in my opinion, my electric water pump adds a nice line of defence against heat surge on switch off but I can't really present any proof its actually necessary. The pump is on the heater circuit as is my LPG vaporiser so for faster change overs to gas in cold weather I just flick it on to get some hot coolant around the vap a little sooner, works well to warm the car's interior too.

Having the electric pump run on with the fan after engine switch off really just came as a secondary idea, and was only implemented because I could.

Dave

Edited by ChimpOnGas on Tuesday 6th December 23:30