Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Higher Voltage: Fool Your Alternator?

Status
Not open for further replies.
Hi,

Yes very interesting thanks.

My rotor coil measures 3 ohms so i wondered if they are PWM'ing the coil rather than direct drive. I mean even for full blast. At 15v 3 ohms would draw 5 amps DC. If the resistance was higher like 12 ohms i would have no problem driving the rotor directly, but because of the low resistance and small diodes i have to wonder.
 
...
My rotor coil measures 3 ohms so i wondered if they are PWM'ing the coil rather than direct drive. I mean even for full blast. At 15v 3 ohms would draw 5 amps DC. If the resistance was higher like 12 ohms i would have no problem driving the rotor directly, but because of the low resistance and small diodes i have to wonder.

When installing alternators in aircraft, I use a 5A (pullable) breaker for the field circuit and a 70A pullable breaker for the B output.

Most 14V Alternators I have played with draw 2 to 4A through the field (rotor) when the engine is not turning. The bang-bang regulators we have shown you all switch the full 12V battery voltage to field when Vout is too low, including when the rotor is not even moving but the ignition key is on.

Since the transistor inside the regulator is either fully-off or fully-on, it does not dissipate much power. The pwm occurs only when the alternator is spinning, and only if the alternator is not maxed out. At high load, low rpm, the transistor in the VR never turns off.
 
Hi again,

Yes that sounds reasonable Mike.

I looked a little more carefully at this and found that at a continuous current of 5 amps the average current though each rectifier will be about 1/3 of that, so 5/3 amp average through one small diode. That's a little more reasonable, so with 3 ohms and a target of 15v we'd see 5 amps average through the coil but much less through each diode. I guess they can take it, and also the rotating current is probably less due to the brushes.

But i also figured out a simple way to get around any worries i might have about trying a direct short to ground to bypass the regulator (temporarily of course). That is, instead of shorting it to ground just use some low value resistance like 5 ohms to start, then see how much that changes things. If it doesnt, then 3 ohms, 1 ohms, etc. That way i dont have to worry about an overcurrent ruining anything.

I'd probably try the diode trick first though, just to see how that works, because that's the simplest to do. I was thinking maybe bring out the two terminals where the diode is to connect after breaking the connection inside (which i still have to verify where it is). That way i can always connect them back together without removing the alternator again or use an external switch to connect it back to normal operation.
Interesting, any series diode will have to handle the full 5 amps continuous.

Any ideas how to bring the wires out, as to what kind of terminals to use?
They have to go through metal that is grounded, so they must be insulated like binding posts going through a metal front panel. They have to survive high heat, and lots of vibration, and possibly some oil contamination on the outside surface.
Or maybe some wires, connected to a terminal strip on the outside of the alternator housing?
It has to be super industrialized so it can handle the harsh environment.
 
There are lots of automotive weatherproof connectors available. Maybe the most famous is the Weatherpak system., but there are smaller ones. Look at Mouser for automotive connectors. The problem is, they usually give you pigtails. Like the MAF sensor, the fixed part of the connector is usually integrated into the body.

You can always use grommets sized properly to get the wires out and if you can somehow fasten an automotive connector securely, you'd be OK, I think. Ty wraps and ty-wrap holder might work.

You might also be able to bring out a pigtail and secure it to a bracket on the engine.

I do have a little story where, the outside temperature sensor was broken and replaced under warranty in a GM vehicle and it broke again, The only real issue was that the sensor was installed using a "flop around" kind of fastener, so it was always wiggling. Securing the wire and the sensor and it's been good for like 10 years or more. The hardest part was finding where the sensor was under the vehicle.
 
Last edited:
Hi again,

I'll have to check out those terminals, but i have a feeling they may be higher priced then they are worth :)
Sounds interesting though, and if anything maybe i can bring out some other terminals too such as that one regulator 'collector' terminal so i can observe what it is actually doing and then modify later.

Guess i'll have to break out the high temperature epoxy too now...
Hey, maybe wires through a blob of epoxy :)
I've checked the exhaust manifold for temperature but never the alternator case. Would be interesting to see how hot that gets on a hot summer day and a 10 mile drive.
 
I doubt the temp would be much greater than I measured somewhere above on a winter day. Here, **broken link removed** is a company that sells tiny quantities of Teflon insulated wire and fiberglass tubing.

I bought some recently to fix a 50 YO dehumidifier. The power cord bit the dust and it was rather special. The strain relief area was slotted and the molded strain relief had to fit in that spot, so I measured the current and went to a local vacuum cleaner store and got a cord by using "Got anything like this"? The exposed ends were too short, so they were made PTFE insulated and placed in a fiberglass tube to protect against abrasion and dry rot. It also got upgraded to have a ground wire. It also got a coat of paint. That took a week to get right.
 
Hi,

Yes some interesting ideas. I'll have to think this all over and think about if it is worth doing.
I'll have to take the alternator out of the car again, work on it, then put it back in again. It's not super hard to do but it's been cold outside around here the past week or so. I am waiting for a warmer day just to put some new battery terminals on :)
Sooner or later i'll do something though, and it will be nice to have it working right again.
 
Hi,

I have some additional information to add now...
I measured a friends car battery and i found it very interesting.

First, the car battery before starting the car (after sitting for only a day or two) measured 12.05 volts. That seems low.
Second, once the car is started, the battery voltage measures 14.55 volts. That seems to match what others have been saying too not only here but in various other places on the web.
Third, the battery voltage jumps right back down to 12.05v when the engine is turned off. This makes the battery look bad.

So the engine running alternator voltage is over 14.5 volts, which seems right, but mine only measures 14.0 volts, which seems too low.

The battery voltage jumps down too fast too once the engine is turned off, so it looks like the battery is not so good. I have read that it should read about 13.5 volts when first turned off (measuring within a few seconds after turn off). Even mine reads something like that maybe 13.1 volts, so his battery looks bad. I told him about this and he said that when he had the battery installed the guy gave him a battery that he did not want because it was a different brand, stating that he could not get the requested brand. This 'new' battery is supposedly only 4 months old now too, and it is a NAPA brand battery. A brand new battery before any use or any charging measures 12.6 volts.
 
I float (no external load) car starting batteries at ~13.3V. On a good battery, If I disconnect the float charger, the battery terminal voltage will slowly drop to ~12.65V (room temperature) over about 24 hours after coming off charge. If a battery dropped to 12.1V or less in that 24 hour period I would be replacing it. See **broken link removed**
 
I float (no external load) car starting batteries at ~13.3V. On a good battery, If I disconnect the float charger, the battery terminal voltage will slowly drop to ~12.65V (room temperature) over about 24 hours after coming off charge. If a battery dropped to 12.1V or less in that 24 hour period I would be replacing it. See **broken link removed**

Hi Mike,

That looks like some good information. I'll have to read it all over.

My battery measures 12.36 volts right now (6:45am) but yesterday (1pm) which was less than 24 hours ago it measured 12.54 volts. So it dropped quite a bit since yesterday already. If i bring it in the house and charge it with my power supply, it will not drop nearly that fast after a day or two later. The day before yesterday it was 12.24 volts. So it did charge somewhat, but i dont believe it charged as much as it should if the alternator was up to 14.5v during charge instead of just 14.0v like it is now.
 
...My battery measures 12.36 volts right now (6:45am) but yesterday (1pm) which was less than 24 hours ago it measured 12.54 volts. So it dropped quite a bit since yesterday already. If i bring it in the house and charge it with my power supply, it will not drop nearly that fast after a day or two later. ...

Might be indicative that the car that the battery is normally in has some parasitic drain with the key off. You might want to measure it. I use my multimeter in the DC mA mode, and I put it in series with with the positive battery lead with the lead disconnected from the battery post.

Over the years, I have seen various parasitic drains on various vehicles. My Cessna has an electric clock that draws about 5mA. My GMC pick-up is about 5mA(keep alive for the ECU and radio, so must be driven every about two weeks). My Ford Motorhome takes the cake. It draws 32mA when parked. It will kill its starting battery stone dead in about a month.

I keep float chargers on all these. They get little use during the winter, and these batteries would suffer due to being chronically discharged with out using the float chargers.
 
Hi Mike,

Wow, 32ma, that's way too much. Wonder what draws that much current in the car.
Mine is 2ma, i dont know what his is yet cause we didnt measure it yet, but that's a great idea, so that's next.
I didnt intend to troubleshoot anybody else's car but since his seems bad too i want to know what is going on with his too now, so the plot thickens :)
I have a volt meter that draws a little too, maybe 2ma. The old volt meter i used was converted into a volt meter from a current meter, so the movement is very rugged, but because it was a current meter with relatively low basic movement resistance it was drawing about 6ma and i thought that was high. It never killed the battery though and never seemed to bother anything even only using the car only once a week.

Thanks again for all your valuable input to this thread. This is part of the 'bread and butter' of life we call transportation, so it's not really in the came category as a hobby it's a necessity, and that makes it very important so more information is always better.
Thanks again to KISS and debe for their contributions also. This stuff could help other readers too down the road.
 
Hello again,

I have some more information to add. This involves some measurements taken over a couple days or so after driving short and longer distances (measurements of my battery voltage). The results are interesting.

First, with the voltage being at 12.24v for 12 hours or more i ran the engine for about 5 minutes. The voltage of course comes up fast to about 14.0 volts, and stays that way for about 5 minutes as the engine runs.
When i come back into the house and after a short time measure the battery voltage again, it reads 12.52 volts.
24 hours later measure the voltage and find it already dropped to 12.24 volts again.
About 12 more hours later, the voltage has dropped to 12.22 volts.

Next, i drive the car for about 10 miles, making various short stops for things like groceries (voltage was 12.22 before starting the engine).
I get home and come in the house and measure the voltage and it measures 12.54 volts, almost the same as last time the engine ran.
24 hours later the voltage has dropped to 12.35 volts, which i consider significantly higher than 12.24 volts especially considering that the start voltage this time was only 12.22 volts.
48 hours later (a total of 72 hours from the last engine run) the voltage measures 12.32 volts.

So we see that the battery does not charge much with only 5 minutes charge on high idle.

To summarize:
Engine off 12.24v, after 5 minute high idle run 12.52v, after 24 hours rest 12.24v, after 36 hours 12.22v.
Engine off 12.22v, after 10 mile drive 12.54v, after 72 hours rest 12.32v.

Very interesting. This is with the older 2009 battery.
The newer NAPA battery i had purchased after a year of use was even worse than this. It would drop below 12v sometimes very quickly. When i brought it back the guy put a battery analyzer on it and said it looked good but still gave me a replacement. This is why i dont trust those quickie battery analyzers.
 
This is why i dont trust those quickie battery analyzers
Ah, but your analysis took 72 hours. I can't see any customer waiting around in a shop that long to get a battery checked :).
 
Ah, but your analysis took 72 hours. I can't see any customer waiting around in a shop that long to get a battery checked :).

Hi,

Well, i didnt mean to imply that they should :)

They could leave the battery and come back once it was accurately tested.
As to where they get another battery in the mean time that is up to them and the place doing the testing.
But using a meter that cant actually test the battery is worse because it gives false results. Might as well pick up a small cardboard box, connect two jumper leads to the sides, and write on the front with magic marker: "This battery is good", and you have your own home made "battery tester" ha ha.
 
But using a meter that cant actually test the battery is worse because it gives false results.
True for your particular battery, but may give better results for others?
pick up a small cardboard box, connect two jumper leads to the sides, and write on the front with magic marker: "This battery is good"
LOL. Perhaps the type of meter the shop was using ;).
 
Battery tests don't tell all. I know that. Not too long ago, I thought the battery was dead because of age. I broke a belt while in Philadelphia. The shop "tested: the battery a month earlier and the other shop did it after the repair. I had to replace the battery the day after the failure miles from home. www.BatteryUniversity.com has some articles on testing.
 
True for your particular battery, but may give better results for others?
LOL. Perhaps the type of meter the shop was using ;).

Hi,

If that first one was true then how do we know what batteries it works for and what ones it does not work for?
The NAPA battery was tested at the NAPA shop. If they dont know what meter to use then who would know :)
 
Battery tests don't tell all. I know that. Not too long ago, I thought the battery was dead because of age. I broke a belt while in Philadelphia. The shop "tested: the battery a month earlier and the other shop did it after the repair. I had to replace the battery the day after the failure miles from home. www.BatteryUniversity.com has some articles on testing.

Hi,

What do you think caused the failure and what was the failure mode?
 
Sitting. Deep-self discharge, And the failure was wierd. I lost belt supposedly. I drove 30 miles. Car was dead after park.. Went to my appt. for an hour. Car started. Went abut 5 miles. Car dead after park. Went to appt. for an hour. Car started. Sixth sense said call for tow. Did. Dx. belt only. Replaced belt. Next morning - No start.

Now about a month prior I asked to have my battery tested and got a go. Car had been sitting for at least 6 weeks.
 
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top