The Museum of HP Calculators

HP Forum Archive 08

[ Return to Index | Top of Index ]

HP65 please help me !!!
Message #1 Posted by Andreas Stockburger (Germany) on 21 Mar 2002, 3:39 p.m.

Hi,

I bought an HP65 on ebay - It is NOT working :-((

Perhaps someone has an idea:

If I connect the calculator with power (stabilized power supply) it takes 150 mA but the display is dark !

So I decided to disassemble the calculator and checked (randomly) some signals with my oscilloscope. Hey! there is live in it! I checked the boards for broken connections - nothing.

I cut the connection to the card-reader to be sure that there is no side-effect here - but nothing.

The calc comsumes 150mA and the display is dark .....

Questions:

1.) should there be dynamically changing sinals at the display-drivers ?

2.) Is 150mA consumption OK (without display) ?

Thanks for your help ..

Andreas

      
Re: HP65 please help me !!!
Message #2 Posted by Ellis Easley on 21 Mar 2002, 9:42 p.m.,
in response to message #1 by Andreas Stockburger (Germany)

I'm no expert, but for a start, see if the DC-DC convertor is working. Look for the toroidal transformer. It is driven by a blocking oscillator and it has windings that are rectified to generate about -12V and +6V (I think) for the MOS chips

            
Re: HP65 please help me !!!
Message #3 Posted by Andreas Stockburger (Germany) on 22 Mar 2002, 1:16 a.m.,
in response to message #2 by Ellis Easley

Hi,

I investigated a little bit further ...

The calculator-electronics itself is working! I can measure fancy digital signals at the display drivers (? -> ICs under the LED-display) The signals changes if I press some keys !!

But the display itself is dark !

Question:

1.) How can I verify if the display drivers are working ?

2.) Whats the pin-layout of the LED-Chips ?

Many thanks for your help

Best regards

Andreas

            
Re: HP65 please help me !!!
Message #4 Posted by Vieira, Luiz C. (Brazil) on 22 Mar 2002, 9:34 a.m.,
in response to message #2 by Ellis Easley

Hi;

Just to add some other interest points to the valuable info from Ellis.

I do not own an HP65, but I am A LOT concerned about these 150 mA with a dark display. It smells like a short circuit (let's take "short circuit" as its definition, not a zero-ohm load), I don't know. I believe 65's circuitry (is the 65's CPU a C-MOS?) would not need that much current, considering no LED is on. That is what we here call a Tupiniquim technique (Tupiniquim is the name of a native indians tribe in Brazil), based on finger touching. If the component is somewhat hot, it is probably the villain.

As you have a Scope, try measuring the waveform at the displays' segments. Should you see the same waveform anywhere, some circuit is opened, right? No waveform and we have no sign for the LED's, which may result from an unexpected load (short circuit?)

Keep us informed. Maybe Tony Duell (as others I don't remember the names now, sorry!) will post something valuable.

Success.

                  
Tupiniquim technique
Message #5 Posted by Ellis Easley on 22 Mar 2002, 6:35 p.m.,
in response to message #4 by Vieira, Luiz C. (Brazil)

The Tupiniquim technique must be used with caution. I used to repair computer boards with 4K dynamic RAMs which, if they were missing the -5V connection, got hot enough to burn their part number markings into your fingertip!

                        
Re: Tupiniquim technique
Message #6 Posted by Tony Duell (UK) on 22 Mar 2002, 9:18 p.m.,
in response to message #5 by Ellis Easley

I remember just that comment ('It's painful to discover the RAM manufacturer's logo burnt into your fingers' or something similar) in the TRS-80 (Model 1) Technical Manual... That dates me somewhat, right...

                              
Re: Tupiniquim technique
Message #7 Posted by Andrés C. Rodríguez (Argentina) on 24 Mar 2002, 4:11 p.m.,
in response to message #6 by Tony Duell (UK)

The reverse technique, used at least in my job in Argentina in the '80s was to use a can of freezing gas (named as "Fault Diagnosing" tool) to cool chips suspected to be more hot than usual, and to see if cooling them makes some change in behavior.

You can burn yourself with the gas if you let it touch your skin for more than a couple of seconds. (I think the gas was not freon, because freon was banned in Argentina from the '70s on, due to ecological concerns. One rare case of good thinking here...)

I would not pretend to name the technique from some indigenous tribe from the Patagonia, the very nice and cold southern part of my country, but, perhaps, ...

(To my good friend Luiz, please don't take this as a criticism!)

                                    
Re: Tupiniquim technique - I agree!
Message #8 Posted by Vieira, Luiz C. (Brazil) on 25 Mar 2002, 8:17 a.m.,
in response to message #7 by Andrés C. Rodríguez (Argentina)

Hello, you all.

Thanks for the comments. Sometimes, we need to think a bit more and practice a bit less.

To my friend Andrés: I would never take it as criticism. The Tupiniquim term sounds too strong, and is somewhat easy to remember. In fact, it reffers to any technique that uses "natural" resource (in this case, hot sensing). I know that, in some cases, even portable, low voltage calculator circuitry can generate enough heat to burn fingers. For this, I'm sorry. (mine have been burnt once or twice...)

Now, a single request. I believe Dave Hicks can help, if so. There are lots of valuable info in here (this forum) about many calculators' HW that could lead to very good, compiled documents. Some calculators, say, HP67/97, 41 series, Voyagers and others, allow user intervention and small repairs. Now, the request: Dave, is it easy to have the Forum Archieves as complete files, instead of links? Say, is it possible having the Forum Archieves downloadable? This way, compiling this or that subject would be easiear. (Is it correct?)

Cheers.

                                          
Re: Tupiniquim technique - I agree!
Message #9 Posted by Ellis Easley on 25 Mar 2002, 5:10 p.m.,
in response to message #8 by Vieira, Luiz C. (Brazil)

I agree that the Forum would be nice to download in bulk or would make a good addition to the CDs. While not every post is memorable, quite a few are real gems.

      
Re: HP65 please help me !!!
Message #10 Posted by David Smith on 22 Mar 2002, 5:27 p.m.,
in response to message #1 by Andreas Stockburger (Germany)

The HP65 draws around 200mA or so with a lit display. 150mA with a dead one is OK. If your display is totally dark, but the machine works, then I would suspect the problem may be in the display chips on the keyboard or the connections to it. The driver chips are basically compatible with those on the HP35 or HP45 and these can be used as a source for parts.

      
Re: HP65 please help me !!!
Message #11 Posted by David Smith on 22 Mar 2002, 7:13 p.m.,
in response to message #1 by Andreas Stockburger (Germany)

Andreas,

I tried to send you an email that describes the HP internal circuitry, but your email filter rejected it. I tried sending it from three different hosts... why bother asking for help if you won't accept it?

            
Re: HP65 please help me !!!
Message #12 Posted by Andreas Stockburger (Germany) on 23 Mar 2002, 1:20 a.m.,
in response to message #11 by David Smith

Hi David,

sorry for the inconvinience with my e-mail at angelfire. Please try this one "andreas.stockburger@gmx.de"

Best regard

Andreas

      
Re: HP65 please help me !!!
Message #13 Posted by Tony Duell (UK) on 22 Mar 2002, 8:47 p.m.,
in response to message #1 by Andreas Stockburger (Germany)

First check the power converter circuit. It produces +6.2V (Vss) and -12V (Vgg) for the PMOS logic chips. The ground reference (for the 'common' lead of you voltmeter or 'scope) is the -ve side of the battery pack. OK, you can find VSS on the 'top' (nearest display) of the connector to the card reader PCB, or on pin 10 of the CPU hybrid. Vgg is on the 4th pin from the 'top' of the testpoints down the opposite edge of the PCB from the card reader connector, or on pin 16 of the CPU hybrid. If those are present, look for a 2-phase clock (square waves) on pins 40 and 41 of the CPU. If missing, then suspect the display anode driver chip. If that clock is OK, then look for a similar pair of clock signals on pins 38 and 39 of the CPU. If those are missing, then the CPU hybrid is faulty :-(. OK so far? Look for activity on CPU pins 11-15. Those carry the display data to the anode driver chip. If there's something there, then the problem is most likely the anode driver chip (on the keyboard/display PCB), the displays themselves (but I wouldn't have thought that all 3 would go simultaneously) or the inductor modules at the top, near the charger connector. But those don't fail often. The display driver is on the on/off swtich side of the display, and is often an 1820-1029. One other thing to check. The cathode driver (1820-1226) needs a +8.2V Vcc supply on pin 10. Check it's there. If not, the display will be blank. It comes from the power converter circuit on the logic board.

            
Thank you all !!! was HP65 please help me !!!
Message #14 Posted by Andreas Stockburger (Germany) on 25 Mar 2002, 3:44 a.m.,
in response to message #13 by Tony Duell (UK)

Hi there,

thank you all for your fantastic help! Tony: the document you sent to me was exactly what I need to fix the calculator !!! I could find a (randomly) broken connection on the "main"-board. It was easy to fix after I found it.

Once again: Thank you all very, very much for your help !!!

Best regards

Andreas

[ Return to Index | Top of Index ]

Go back to the main exhibit hall