HP-50G Excessive Battery Drain / Fix?

[HrastProgrammer]

(06-03-2015 01:48 AM)rprosperi Wrote:  Very jealous here... I've wanted to attended these, but the closest they come to the NY area usually is Las Vegas. I think this race will be covered on TV here this weekend. Who won?

Hannes Arch won, the same as last year. Paul Bonhomme was disqualified just a few seconds before the end of his lap because he exceeded 10G. This is a very demanding track because of the winds which regularly change direction between the training, qualifications and the race. We were very lucky to have the race second year in a row here, as those small adriatic towns are not really suitable for such rather large events.

Beside the race (and the fact that my wife likes Rovinj very much, we live just 250km from there) my main goal was to make a lot of sound recordings (on the airport itself) which I could eventually use in my own music. And seeing that "Flying Bulls" Lockheed P-38 Lightning in a hedgehop just a 50m above us was really fascinating

[Garth Wilson]

I just checked mine. I got 18mA idle, 5.7mA off. I knew it was bad, because the first set of brand-new alkaline batteries was dead and leaking after only a month or two. I had fiddled with it some when it was given to me, but the manual it came with was frustrating and useless, and I put the calc in a drawer.

[Frido Bohn]

(06-03-2015 05:09 AM)Garth Wilson Wrote:  I got 18mA idle, 5.7mA off.

This is quite interesting: in all three examples with high basal (off state) battery drain, the basal current adds up to current at the idle state. Is this a common trait that may help explain how this defect works?
A well working HP50g should consume about 12 mA in the idle state, and only a few micro Amps in the off state.

Cheers,
Frido

[jebem]

It was interesting to know what are the manufacturing year of the affected machines.
Perhaps this can be related to a specific production batch?
Can you post your serial numbers for comparison?

So, I took some additional readings from my two 50G machines.

This time I've used my vintage lab precision analog current meter by SIEMENS Multizet Model M819-9.1B-015 just for fun!
(For power on state readings I used the 100mA range (2.5Ohm internal resistance)

Machine 1 - New old stock, original firmware:
HP-50G s/n: CNA 90302447 (year 2009)
Version HP50-C Revision #2.08 HP 2006
Power off state: 32uA
Power on, idling, clear screen: 14.8mA

Machine 2 - well used, updated firmware one year ago after buying it:
HP-50G s/n: CNA 02109127 (year 2010)
Version HP50-C Revision #2.15 HP 2009
Power off state: 27uA
Power on, idling, clear screen: 12.8mA

These readings didn't change when:
- Having a 1GB SD card inserted or not;
- Having the battery backup installed or not;

[Garth Wilson]

(06-05-2015 07:34 PM)jebem Wrote:  It was interesting to know what are the manufacturing year of the affected machines.
Perhaps this can be related to a specific production batch?
Can you post your serial numbers for comparison?

The one I told about above, with 5.7mA "off" current, is CNA 70702169, and nothing was plugged in, including no SD card.

[Tugdual]

Just for the records, mine is:
Off: 23 μA
Idle: 12.8 mA

Not using SD Card, it delays boot sequence too much.

[cruff]

The measurements for mine (CNA9470....) is

off: 29.5 µA
idle: 12.5 mA

[rprosperi]

(06-03-2015 05:06 AM)HrastProgrammer Wrote:  Hannes Arch won, the same as last year. Paul Bonhomme was disqualified just a few seconds before the end of his lap because he exceeded 10G. This is a very demanding track because of the winds which regularly change direction between the training, qualifications and the race.

... And seeing that "Flying Bulls" Lockheed P-38 Lightning in a hedgehop just a 50m above us was really fascinating

There was a similar disqualification at the race a week before in Japan - exceeding 10G in the final pylons. I wonder how that feels ?

Yeah, warbirds on the deck certainly do raise one's blood pressure. I believe the FB Lightning is a polished aluminum finish, right? I've never seen this warbird live (it is in Europe - I'm not). I was lucky enough to attend "The Final Gathering" in 2007, where just under 100 Mustangs participated, plus just about every other type of WWII warbirds, including 3 Lightnings. Doubt I'll ever see that again.

Thanks!

[HrastProgrammer]

(06-06-2015 12:29 AM)rprosperi Wrote:  There was a similar disqualification at the race a week before in Japan - exceeding 10G in the final pylons. I wonder how that feels ?

Must be disappointing, especially if you had the fastest lap until this point. He just pushed it too hard ...

Quote:Yeah, warbirds on the deck certainly do raise one's blood pressure. I believe the FB Lightning is a polished aluminum finish, right?

Yes, it is.

http://www.flyingbulls.at/#/de/Aircraft/Lockheed-P-38L/

It is located at the Salzburg airport in Austria, if I remember correctly.

Quote:I've never seen this warbird live (it is in Europe - I'm not). I was lucky enough to attend "The Final Gathering" in 2007, where just under 100 Mustangs participated, plus just about every other type of WWII warbirds, including 3 Lightnings.

Now I am jealous! That must have been a fantastic feeling ...

I wanted to be a pilot back in the day but, unfortunately, some hearing problems destroyed this dream and I had to stick with MS Flight Simulator

[jebem]

I have experimentally confirmed that the current consumption from the 4 x 1.5V AAA dry batteries will be 0.0mA when plugging the calculator to a 5V USB power source.

Every time I plugged the USB cable to my PC, the dry batteries current dropped to zero mA. Unplugging the USB cable, the current rises right away to the normal value.

Smart power supply mechanism inside the 50G. And it means that it really work nicely with only 5VDC.

That explains why a lot of people posted here in the old forums about using 1.2V rechargeable batteries without problems at all.

[jebem]

(06-05-2015 10:16 PM)Garth Wilson Wrote:  

(06-05-2015 07:34 PM)jebem Wrote:  It was interesting to know what are the manufacturing year of the affected machines.
Perhaps this can be related to a specific production batch?
Can you post your serial numbers for comparison?

The one I told about above, with 5.7mA "off" current, is CNA 70702169, and nothing was plugged in, including no SD card.

So far, the affected calculators shows older manufacturing dates. Yours should be from 2007, right?

I wonder if this issue is related to older firmware releases, that didn't put the processor in deep sleep, for instance.
My two calculators were made in 2009 and 2010 and runs two different firmware versions: 2.08 and 2.15. Both are fine and don't exhibit this current "leaking".

Could you check your software version, please?

[Garth Wilson]

(06-06-2015 04:27 PM)jebem Wrote:  

(06-05-2015 10:16 PM)Garth Wilson Wrote:  The one I told about above, with 5.7mA "off" current, is CNA 70702169, and nothing was plugged in, including no SD card.

So far, the affected calculators shows older manufacturing dates. Yours should be from 2007, right?

I wonder if this issue is related to older firmware releases, that didn't put the processor in deep sleep, for instance.
My two calculators were made in 2009 and 2010 and runs two different firmware versions: 2.08 and 2.15. Both are fine and don't exhibit this current "leaking".

Could you check your software version, please?

To be frank, I don't even know how to find out. The 50g was given to me hardly used, and the tiny 1/4"-thick manual that came with it was pretty worthless and doesn't even have an index, and I don't see anything in the table of contents that would tell me how to check the software version. I already had such an investment in the 41cx, 71, and other computers that I pretty much had what I needed and was not motivated to print out the much better manual online and put the time into learning the 50g. I've offered it to a couple of relatives who are mechanical and electronics engineers, and for the same reason of the apparently very large investment in time required to learn to really use it, they were not interested. If you can tell me an easy way to check the version, I'll do that and report back.

[rprosperi]

(06-06-2015 06:25 PM)Garth Wilson Wrote:  

(06-06-2015 04:27 PM)jebem Wrote:  So far, the affected calculators shows older manufacturing dates. Yours should be from 2007, right?

I wonder if this issue is related to older firmware releases, that didn't put the processor in deep sleep, for instance.
My two calculators were made in 2009 and 2010 and runs two different firmware versions: 2.08 and 2.15. Both are fine and don't exhibit this current "leaking".

Could you check your software version, please?

To be frank, I don't even know how to find out. The 50g was given to me hardly used, and the tiny 1/4"-thick manual that came with it was pretty worthless and doesn't even have an index, and I don't see anything in the table of contents that would tell me how to check the software version. I already had such an investment in the 41cx, 71, and other computers that I pretty much had what I needed and was not motivated to print out the much better manual online and put the time into learning the 50g. I've offered it to a couple of relatives who are mechanical and electronics engineers, and for the same reason of the apparently very large investment in time required to learn to really use it, they were not interested. If you can tell me an easy way to check the version, I'll do that and report back.

The serial number (sticker) on the back, if still there, will reveal the mfr date.

For the firmware rev, type VERSION. It will reply with several strings, including IIRC likely either 2.08 or 2.15. 2.15 is the most current.

[Garth Wilson]

The serial number (given above) is CNA 70702169.

I don't see how to scroll to the right far enough to see anything meaningful in the version response. It says:

{Version HP-50-C[ENTER-type arrow]Revis[right-arrow]

[Joe Horn]

(06-07-2015 12:31 AM)Garth Wilson Wrote:  The serial number (given above) is CNA 70702169.

I don't see how to scroll to the right far enough to see anything meaningful in the version response. It says:

{Version HP-50-C[ENTER-type arrow]Revis[right-arrow]

Either press up, up, VIEW (the F2 key), or change to RPN mode and make sure flags -52 and -65 are both clear (to allow level 2 to be displayed on two lines).

[robert rozee]

would someone with a good HP-50g please try the following: while measuring the current consumption, and with the calculator turned off, press various keys on the keyboard and see if any cause the current consumption to increase.

cheers,
rob :-)

[Garth Wilson]

Ok, I got 2.08, copyright 2006.

[Joe Horn]

(06-07-2015 02:44 AM)Garth Wilson Wrote:  Ok, I got 2.08, copyright 2006.

Good heavens, upgrade to 2.15! --> http://www.hpcalc.org/details.php?id=7097

[HrastProgrammer]

(06-06-2015 04:27 PM)jebem Wrote:  So far, the affected calculators shows older manufacturing dates. Yours should be from 2007, right?

I wonder if this issue is related to older firmware releases, that didn't put the processor in deep sleep, for instance.

This problem is not related to the firmware because my piece of chinese crap is CNA00301073 (this should be year 2010, right?) and has ROM 2.15 installed. It's a fault on the PCB and it could be that a CPU doesn't enter a deep sleep when turned off (edit: this doesn't explain 24mA idle current in my case) but I doubt there is anything we can do about it.

After 10 years of ignoring all new HP calculators (my "cash register" 49G+ from 2003. is the biggest piece of electronic junk I have ever seen) I had a chance to work with 50G and I actually liked it to some extent - bigger display and 2x speed compared to 48G, and a better keyboard than 49G+, work very well for my HRAST BASIC. But this (obviously rather common) problem really pissed me off.

That said, I will buy another new 50G (so I can have it fixed under warranty if needed) because I can get it for cheap and I need a calculator "to go" but I didn't change my opinion about it that I wrote almost 12 years ago:

http://www.hrastprogrammer.com/hp49gp.htm

And my 4x48GX certainly don't go anywhere, especially after I opened a 50G because it doesn't look like made by HP at all. I have an opened 48G+ side by side and they are like night and day from inside (and outside, for that matter) ...

[jebem]

(06-07-2015 02:43 AM)robert rozee Wrote:  would someone with a good HP-50g please try the following: while measuring the current consumption, and with the calculator turned off, press various keys on the keyboard and see if any cause the current consumption to increase.

I tested with one of my good condition HP-50G machines from year 2010, f/W Revision #2.15.
Nope, the current consumption doesn't change from the 27uA while pressing keys when the machine is powered off.

The only exception is the ON key, of course!