Alternator problem

1993 900S turbo convertible (USA)
Simple question, but complicated lead-in, so that you may understand where I have been. Let me open by saying that I am electrical engineer.
I have had 3 Saab 900's, all of the same vintage. Only the one above is left. I love these vehicles, but they all have shared one problem - they eat batteries. On each of them, the day comes, repeatedly, and sooner rather than later, that when I go to start it, the battery is flat for no apparent reason. I've done all the standard stuff - checked the grounds, checked the connections throughout, maintained the batteries properly, looked for intermittent parasitic drains (stuck switch, etc). I replaced the voltage regulator with a new Bosch unit. I checked the alternator terminal voltages in accordance with Bentley. No apparent problems, and no fixes. As a desperation measure, I ran separate wires (#10) directly from the alternator output terminals to the battery terminals, and that helped. But not enough.
Recently, the thing stranded my wife. Again. And I got the ultimatum. Fix it or sell it. So I got more seriously interested in what's going on. Instead of spot checks, I attached a permanent voltmeter to the cig lighter circuit, and drove it for several days. And lo and behold, the problem quickly became obvious. The alternator is inadequate to the task!
With the air conditioning off, and no big drains (engine cold so radiator fans off, power windows/seats quiet), the voltage stays well up, in the range of 13.8-14.0 volts. But as soon as a big drain shows up, the voltage drops to the 12.5 region, and that is not enough to maintain a battery.
So I checked the alternator specs in Bentley. The "80 amp alternator" puts out 80 amps only when the engine is screaming at 6000 RPM! And I don't drive that way (that's why I still have a 1993 Saab!) At 2000 RPM, it is spec'ed to put out only 54 amps. And below 1900 RPM, the output is essentially zero. And that is pretty pitiful against the requirements of the electric radiator fans and the air conditioner clutch. In these conditions, I can actually see the voltage drop and rise as the directional signals flash. And Lord knows what happens if I turn on the headlights.
So. Now I know what is happening. The alternator cannot put out sufficient current to support all the goodies that Saab has installed. And thus the final question (thanks for your patience) is whether there is a way to increase the alternator output at low RPM - OR - is there an aftermarket alternator that will put out a bit more current at low RPM, without making major mechanical mods to the alternator mounting system.
Bill
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 01 May 2008 18:14:25 -0700, Bill Jeffrey

Are you sure the RPM figures you've found are for engine RPM and not alternator RPM? There's normally some gearing done on the pulleys. It's about 2.8:1 on my 2001 9-3. 2000 alternator RPM is well below the engine idle speed.
It certainly sounds wrong that you can't maintain a charge with the engine running though. The 9-3 happily manages 14.4V at idle with headlights, foglights, HRW and seat heaters all on. You could get the alternator tested off the car if you take it to an auto electrical specialist. If it really is bad design, you could try fitting a smaller alternator pulley, but be careful that the alternator won't get over-speeded at max engine revs. Do you know what the pulley ratio is on the 900?
The other way would be to fit a higher capacity alternator from a different model - they do tend to work better all the way down to low revs. Check pulley ratios again though. Bigger alternators are often intended to run faster.
Cheers,
Colin.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

In my both my running 900's (81 turbo and 89 900i), the big increase in load when the fans turn in is quite noticable, esp. if the car is idling. In the 89 900i especially, it creates so much load that the engine speed is noticably affected for a few seconds. I doubt the alternator is doing it directly though - would be more the reduction in voltage affecting the EZK ignition box.
I have a voltage gauge (connected not directly on the battery but on the switched side of the power to the interior devices so it shows the real voltage) in my 81 turbo and when the fans turn on at idle, the voltage temporarily goes down to around 8.5 volts but quickly comes back up to 12.5 or so. I haven't tried watching the voltage level in the 89 car when it's fans turn on.

80A isn't really much considering all the potential loads that are present, but remember most of this is 'peak' load. All the base-load electrical power is actually from the battery, and the alternator just handles peak loads while also recharging the battery.
Things like the fans, headlights, etc. are not base-load electrical devices so take them away (not that you really can, but if you could) and the alternator wouldn't have any problem at all.
I've noticed that both alternators in my 900's have mounting bushes that are a bit worn so they're in need of repair, and the alternators themselves don't sound the best, so they probably are not giving optimal output. They do both have new voltage regulator units though.

Hmmm, interesting point. The alternator pulley size is about half to 1/3 the crankshaft pulley size in my c900's (not sure if this is the same in other Saab models with different belting arragements) so that gives an automatic 'gearing' by virtue of pulley sizes alone of somewhere between 2:1 and 3:1. The alternator at idle (nominal 850 rpm for a c900) is rotating at approximately 1500 to 2000 rpm (if my basic math on how pulley size affects driven devices is correct!). This shows why things like belt tension, etc. are important as the bearings must get a thrashing in alternators.

I think a good thing to check would be the voltage regulator in the OP's car alternator. If the brushes are worn down, or the v/reg itself is fairly old, it might not be able to provide the proper 13.5 to 14.5 nominal DC output under the full range of normal rotational speeds.
That relates back to my point before about how the alternator is a 'peak' load device. It's primary function is charging the battery, and the peak load 'feature' is secondary, though because when charging it's pushing electrons against the battery's internal resistance, much of the electrical output from the alternator feeds directly to load devices even though that's not the main function of the alternator.

That's possible. I don't think Saab used serpentine style belts on any c900's so that could limit the choice as high-amperage alternators all seem to be serp-belt types. Standard double-belt pulley alternators are fairly common on places like Ebay, etc. but I haven't come across many that have ratings of more than about 80 or 90 A. That might affect what the OP could achieve with a replacement (non-standard) alternator to replace the normal one.
Craig.
--
Craig's Saab C900 Page at | Craig's Classic Saab Workshop - Sydney .au
http://lios.apana.org.au/~c900 | http://www.classicsaab.net and other URL's
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Fri, 2 May 2008 23:13:37 +0000 (UTC), Saab C900 Viggenist

It's possible that the ignition is being upset, but it's more likely that the alternator really is slowing the engine down mechanically. At a guess, the rad fan might pull 50A or so as it starts. That's getting on for 1 BHP, which is significant for an idling engine. In the 80's cars often had "idle-up" systems. A solenoid valve in the carb would increase engine power slightly if any big electrical loads (or aircon) were turned on. These days, the idle speed has closed-loop control.

[snip]
That's not right. The battery can't provide any power at-all when it's charging. During normal running, the alternator provides *all* of the power, with extra to charge the battery. If a peak load happens that the alternator can't manage, the voltage drops to the point where the battery stops charging and starts supplying the deficit. The voltage drop when this happens is quite noticeable. Normally these peaks are very short - maybe 0.5 seconds or so for the rad fan starting. If they happen for long periods (say, all the time the rad fan is running at idle) then the danger of flattening the battery starts to increase.
Cheers,
Colin.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Gentlemen -
Thank you for your replies. I'm digesting them, and will let you know if anything develops.
Colin - no I can't be not sure if the stated RPM figure is for engine RPM or alternator spindle RPM. I assumed it was engine RPM, for a couple reasons. First, max cited speed is 6000 RPM. This is more-or-less red line for the engine, so it makes sense to have a spec there. If it meant alternator RPM, that would be only 2000 RPM (engine), which is only low cruise speed. However, I do note that 6000 RPM (engine) means 18,000 RPM for the alternator, which seems kinda high for a mass-produced commodity metal item. Bottom line, I don't know.
I would love to fit a higher-capacity alternator, but space is at a premium in that area. I'm not at all sure that I could wedge a physically-larger unit in there. I was hoping someone had done so, and I could leech on that knowledge.
Viggenist - all of my 900's have exhibited exactly the same stumble and drop in RPM when a big load comes on at idle. In particular, when the radiator fan comes on (you can hear it, of course), the engine very nearly stalls until the control mechanism, whatever it is, brings the throttle up. And when the battery gets close to dying, the engine does, in fact, stall when this happens.
I'm sure that the brushes are in good shape, since they are part of the voltage regulator assembly, which I replaced a few thousand miles ago.
I have confirmed that the problem is (almost) entirely load-related, and not as RPM-related as I had thought. Yesterday I pulled into the driveway and sat there with the engine idling. Voltage measured around 13.5, which is adequate but not generous. When the rad fan came on, and after the stumble recovery, it was down around 12.8, which is not adequate. By adding headlights, and then air conditioner, and then passenger compartment fan, I could pull the voltage down into the 11's. HOWEVER, the interesting thing is that blipping the gas didn't seem to bring it up appreciably. I'll have to ponder this.
Again, thanks for your thoughts and ideas.
Bill
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sat, 03 May 2008 08:39:54 -0700, Bill Jeffrey

All the alternators I've taken an interest in (which isn't all that many to be honest) have been specced in alternator RPM. It makes sense since that makes it independent of the pulley ratio. 18000 RPM seems a lot compared to the engine, but there's no pistons or anything so it's all just rotation. The 9-3 definitely has a roughly 2.8:1 pulley ratio, so its alternator really does get close to 18000 RPM on the red line.

The more I think about it, the more I think this can't possibly be bad design. Even in th 80s, no-one would have accepted a car with a charging system anywhere near as bad as you describe. Therefore yours must just have a fault and it doesn't need a different alternator, just a working one (or working wiring).
Getting it tested off the car would be my next step, or put your electrical engineering skills to good use and make an alternator ammeter. A cheap multimeter, set to 200mV and wired across the main output cable should do. You'll need experiment to work out a multiplication factor to make the readings meaningful, but it should easily get you a semi-accurate idea of how many amps the thing can manage before the voltage starts to drop.
Cheers,
Colin.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Get an AL129X 115 amp alternator from eeuroparts.com or any local supplier. It fits; I know because I have one in my 86 900. Also, you can install an adjustable voltage regulator on your existing alternator. Try:
http://davebarton.com/AdjustableVoltage.html
Which would be considerably cheaper if it works. (I have not tried this.)
Don Kirkpatrick Also an electrical engineer

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

You would be better served to attach the volt meter as close to the positive post of the battery as possible. The inside wiring is going to have some voltage drop across it under heavy load but that doesn't meant the voltage at the alt. or the battery has dropped.
I would think you are barking up the wrong tree thinking you have a design flaw. Thousands of people have been driving thousands of these cars for a number of years. If they had such an obvious design fault it would be well documented by now.
For temporary relief buy yourself one of those small jump start boxes. They have an internal battery.. you hook them right to your battery and start the car. I didn't think they would be of much use myself until a family member got me one and now I have one if every car.
Steve B.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Steve B. wrote:

Steve -
I'm not calling it a design flaw, for the very reason you suggest. What I am suggesting is that Bosch is a bit conservative about setting the voltage regulation. And since there are tolerances on any device, some units come in a bit lower than conservative. The good news is that a Bosch will never destroy a battery by overcharging it - but some will be on the low side, and that hurts batteries, too.
As for well documented? Well, I've been fighting this problem for more than 10 years. In the course of that time, I've monitored this newsgroup, as well as a couple owners' discussion boards, and this problem of low battery comes up again and again. It also comes up on Volvo and Mercedes Benz groups. Most people don't have the problem - but a few do. It reminds me a bit of Lucas electrics on British cars - the Prince of Darkness is not solely a British invention.
Now, when it happens, you can lay it off to bad wires, bad grounds, bad batteries, bad connections, owner neglect, or anything else of the sort. But most of the people who describe the problem also describe the measures they have taken to identify and remedy those faults - and almost always without noticeable improvement. Although there may be doubt in your mind, there is no doubt in mine that the problem does occasionally occur. I'm not accusing Bosch of anything. All I would like is a bit of help in my own particular situation.
Thanks for your thoughts.
Bill
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Motorsforum.com is a website by car enthusiasts for car enthusiasts. It is not affiliated with any of the car or spare part manufacturers or car dealers discussed here. All logos and trade names are the property of their respective owners.