NiZN and the Compass

[+appalachiancachin]

Hi,

 

I recently bought some NiZN PowerGenix batteries for my 62s and since I installed them my 3-axis compass has been acting goofy. By goofy I mean sticking on a certain direction no matter how I turn, then there are times I will move along a little bit and it seems to work OK. Could this be due to the battery change? Has anyone else seen anything like this?

 

Thanks!

[+dfx]

Did you actually recalibrate the compass?

[+ecanderson]

Hi,

 

I recently bought some NiZN PowerGenix batteries for my 62s and since I installed them my 3-axis compass has been acting goofy. By goofy I mean sticking on a certain direction no matter how I turn, then there are times I will move along a little bit and it seems to work OK. Could this be due to the battery change? Has anyone else seen anything like this?

 

Thanks!

After you install those, you MUST recalibrate. The voltage is higher than what your device has seen from previous cells that you have installed. It's another stake in the heart of other theories about why Garmin's compasses drift. It IS voltage.

 

So let's take it from the top, and with luck, you'll never worry about calibration again:

 

Top off a pair of your NiZn cells.

Install them in your GPS.

Turn on GPS.

Wait about 5 minutes.

Perform the calibration procedure.

 

When your cells drop the first bar on the indicator, you have at most 15 minutes to exhaustion. Go ahead and swap them with a fresh set as soon as you see that bar. Doing this will prevent you from needing to calibrate again, and that calibration would only last for a few minutes anyway.

 

Be advised that PowerGenix gave up on trying to penetrate the consumer market with their NiZn technology, and have directed all of their efforts toward automotive and similar batteries. So get 'em while you can find 'em if you like the way they perform in your unit.

[John E Cache]

By goofy I mean sticking on a certain direction no matter how I turn

The whole point of the magnetic sensor is so the GPS knows which way the GPS is pointed. The arrow is supposed to stick on the direction of the cache no matter which way you turn.

[jimlarkey]

Be sure to take another backup set of batteries, such as eneloopes...I've had 3 of the 4 NiZn Powergenix die on me...won't hold charge. Also, Powergenix has not answered their customer phone support since last December. As a previous poster observed, they've moved on, and so have I.

Good luck Buddy!

[+PokerLuck]

Powergenix no longer makes NiZn batteries, though you can still find them with some retailers. They were great batteries, but they had one very fatal flaw. If a cell was ever discharged below 0.9 volts, it suffered permanent damage and could not hold a charge after that. Because they're used in pairs in most GPSs, one of the cells will always be a little stronger than the other and will drive the weaker cell down in voltage. Since they start with such a high voltage, the electronics will be quite happy even with one of the cells below 0.9 volts. I held high hopes for NiZn, but I've switched back to NiMH.

[+kwcahart]

Ditto PokerLuck. Me too, just had too many goofy things happen with the NiZn batteries, I'm now using regular alkaline until my new 2700MAH eneloopes and maha charger get here. And no, the compass is not supposed to stay the same way no matter which way you turn. It should turn towards whatever you are navigating to, and mine was doing the same thing as the OP. If I walked a few yards it would get itself back together and point to the cache or waypoint, but it would stick sometimes just like the OP described.

[+ecanderson]

By goofy I mean sticking on a certain direction no matter how I turn

The whole point of the magnetic sensor is so the GPS knows which way the GPS is pointed. The arrow is supposed to stick on the direction of the cache no matter which way you turn.

I suspect what he's saying is that the arrow 'sticks' such that it remains in the same position on the face of the unit, even when you turn the unit to another bearing... essentially becoming unresponsive for some period of time. I've seen that, too.

[+ecanderson]

If a cell was ever discharged below 0.9 volts, it suffered permanent damage and could not hold a charge after that.

I maintain that the best way to use these is to wait until the top bar on the Garmin 'gas gauge' drops out, then swap the cells immediately. Not only prevents the problem you mention, it avoids any recalibration - ever. You've only got a few minutes of use remaining once that first bar drops anyway. These things have an exceptionally flat discharge profile - right up to exhaustion. Edited September 18, 2011 by ecanderson

[John E Cache]

Hi,

 

I recently bought some NiZN PowerGenix batteries for my 62s and since I installed them my 3-axis compass has been acting goofy. By goofy I mean sticking on a certain direction no matter how I turn, then there are times I will move along a little bit and it seems to work OK. Could this be due to the battery change? Has anyone else seen anything like this?

 

Thanks!

After you install those, you MUST recalibrate. The voltage is higher than what your device has seen from previous cells that you have installed. It's another stake in the heart of other theories about why Garmin's compasses drift. It IS voltage.

I thought I would respond to your comment now that you have pointed out the battery voltage drops with use. That is the main reason why am leaning towards the idea that voltage is a Red Herring and has nothing to do with calibration.

 

My alternate theory you mention is that the doped semiconductor used in the sensor changes when an electrical charge is passed through it. The amount of charge in one battery charge ages the semiconductor enough to warrant recalibration.

 

Also, I have heard "goofy" and "erratic" used to describe normal magnetic compass behavior. In fact, I used those words before I figured out how the compass works.

[+ecanderson]

The amount of charge in one battery charge ages the semiconductor enough to warrant recalibration.

If that were true, these components wouldn't last very long! More to the point, if you see the problem begin, and put in a pair of cells that match the voltage that was present during the most recent calibration, the operation returns to NORMAL -- nothing is changing about the characteristics of the chip over the course of a few hours. What's changing is the externally applied voltage. Keep giving that chip the same voltage and it won't mess with you.

 

The part that Garmin is using is apparently not one of those that uses an internal voltage reference for the A>D off what amounts to an internal Wheatstone bridge, and either the chip's analog portion is being asked to run off Vcc, or there's no decent regulation on the analog voltage reference. No good reference voltage, no good output.

 

Further, the sensor elements are likely NOT made of a "doped semiconductor" material. The newer chips are usually made of Permalloy (a nickel-iron material) or some other magnetoresistive material (ala Honeywell, PNI, ALPS and Memsic, etc.). The old indium components are for older design Hall sensors (ala AKM).

[John E Cache]

The amount of charge in one battery charge ages the semiconductor enough to warrant recalibration.

If that were true, these components wouldn't last very long! More to the point, if you see the problem begin, and put in a pair of cells that match the voltage that was present during the most recent calibration, the operation returns to NORMAL -- nothing is changing about the characteristics of the chip over the course of a few hours. What's changing is the externally applied voltage. Keep giving that chip the same voltage and it won't mess with you.

 

The part that Garmin is using is apparently not one of those that uses an internal voltage reference for the A>D off what amounts to an internal Wheatstone bridge, and either the chip's analog portion is being asked to run off Vcc, or there's no decent regulation on the analog voltage reference. No good reference voltage, no good output.

 

Further, the sensor elements are likely NOT made of a "doped semiconductor" material. The newer chips are usually made of Permalloy (a nickel-iron material) or some other magnetoresistive material (ala Honeywell, PNI, ALPS and Memsic, etc.). The old indium components are for older design Hall sensors (ala AKM).

OK, I admit I was trying to poke you a little for fun. You haven't convinced me, though.

 

The voltage argument leads to some pretty weak conclusions, in my opinion. I can not believe the GPS has no internal regulated voltage, for instance. The speed of CMOS varies with voltage. Speaking of CMOS, the p transistors in CMOS logic also degrade over time.

 

I have always wondered why calibration? When I read an article about Hall Effect chips and how they need to be calibrated periodically, I started discounting the voltage theory. The voltage theory really started crumbling when my new Android phone started asking me to calibrate my compass periodically.

 

Wheatstone bridge? Cmon, man. You can do better than that. :-)You keep bring up other sensor types, but never say if they need to be calibrated. Do you have an example product that uses these parts?

[+ecanderson]

OK, I admit I was trying to poke you a little for fun. You haven't convinced me, though.

 

The voltage argument leads to some pretty weak conclusions, in my opinion. I can not believe the GPS has no internal regulated voltage, for instance. The speed of CMOS varies with voltage. Speaking of CMOS, the p transistors in CMOS logic also degrade over time.

 

I have always wondered why calibration? When I read an article about Hall Effect chips and how they need to be calibrated periodically, I started discounting the voltage theory. The voltage theory really started crumbling when my new Android phone started asking me to calibrate my compass periodically.

 

Wheatstone bridge? Cmon, man. You can do better than that. :-)You keep bring up other sensor types, but never say if they need to be calibrated. Do you have an example product that uses these parts?

The component degradation theory (or any other apart from voltage, for that matter) is simply blown out of the water by replacing cells at lower voltage with cells at the original voltage. When cells are installed that are of a voltage equal to that of the cells during the last calibration, everything returns to normal. I don't know how the experiment could be made any easier.

 

I can not believe the GPS has no internal regulated voltage, for instance.

I'm quite certain that there is internal regulation in these units. The range of operating voltage alone tells us that -- everything from about 2.3V (may be less, I can't recall) or so up to 3.6 seem to be tolerated well enough. Whether there is correct or adequate regulation to the compass chip isn't determined by the regulation (perhaps at two or more different voltages) for the rest of the board. Personally, I'm not inclined to crack open my perfectly functional Oregon to find out whose chip is actually being used, nor to monitor the voltage to it (assuming the package even permits that) during a deep battery cycle, but I might consider doing so on my Dakota 20 some day.

 

Edit: Forgot to respond to this bit:

Wheatstone bridge? Cmon, man. You can do better than that.

Study up on how a magnetoresistive sensor works. It really IS used as the leg of a bridge, with the output voltage of the bridge representing the sensor output. Edited September 22, 2011 by ecanderson

[+kwcahart]

I used the words "goofy" and "erratic" AFTER I learned how to use my GPSr and compass. Maybe you should learn a little something before you make your dumb*** remarks.