The Museum of HP Calculators

HP Forum Archive 04

[ Return to Index | Top of Index ]

Display Power Up on a Classic
Message #1 Posted by Mike Davis on 2 Nov 2000, 12:49 p.m.

I think I read somewhere that when you turn on a classic that the display should "just" show 0.00 and not flash all the digits (as if running a program).

Is that true?

Does it apply to all of the 35, 45, 55, 65 and 67s or just some of these models?

If it does flash, does that point to any specific problem?

Can this happen and still mean there is nothing wrong with a calculator?

      
Re: Display Power Up on a Classic
Message #2 Posted by Viktor Toth on 2 Nov 2000, 7:22 p.m.,
in response to message #1 by Mike Davis

> when you turn on a classic that the display should "just" show 0.00

Correct.

> and not flash all the digits

Most certainly not.

> Does it apply to all of the 35, 45, 55, 65 and 67s

Yes, without exception (except that the HP-35 displays 0. because it has floating decimals.)

> If it does flash, does that point to any specific problem?

No, it can be anything.

> Can this happen and still mean there is nothing wrong with a calculator?

Yes, if something is wrong with the power. For instance, if the calculator gets unregulated DC from a power supply, it may display random digits. Mind you, if it is subjected to that power input for any length of time, chances are that it will develop a permanent problem even if it was working fine previously.

Viktor

            
Re: Display Power Up on a Classic
Message #3 Posted by Mike Davis on 2 Nov 2000, 9:53 p.m.,
in response to message #2 by Viktor Toth

Thanks,

I'm wondering if it has something to do with the power switch not making good contact.

I notice the following on a 45: 1) When the switch is switched on, the display looks like the display on a 67 when it executes a short program that returns 0.00. It seems to flash all digits (not random but similar to when a 67 runs.

2) I also notice that when everything is ok, just barely touching the power switch makes the display act a little odd. IF the switch is pressed in a little, the display becomes consistently good.

This second problem makes me think the calc needs to be opened up and the power switch cleaned. I wonder if this could also be the source of the display flashing. Like maybe the switch has some bounce that messes up the processor or something. Hmmm...

                  
Re: Display Power Up on a Classic
Message #4 Posted by DaveJ on 2 Nov 2000, 10:43 p.m.,
in response to message #3 by Mike Davis

The switches can get dirty. HP used a conductive grease to improve contact and if this grease gets grimy it can cause switch problems. It is easy to clean once you get the calc open and past the issue of label removal risk...

                        
Display Power Up --> Switches
Message #5 Posted by Katie on 3 Nov 2000, 12:54 a.m.,
in response to message #4 by DaveJ

I'm curious as to what other people here do to clean the switch contacts. I've simply been removing the gritty grease entirely and cleaning the contacts. I haven't found a grease to replace what HP was using and I don't think that it's really necessary for good contact (although the switch does have a better feel to it with the "clean" grease in there).

                              
Re: Display Power Up --> Switches
Message #6 Posted by r. d. bärtschiger. on 3 Nov 2000, 9:10 p.m.,
in response to message #5 by Katie

I remove the old grease then apply a thin coating of 'gc 568', which I believe is a silicon grease. Gc is the brand. Should be available at electronics stores.

rdb.

                  
Re: Display Power Up on a Classic
Message #7 Posted by Viktor Toth on 3 Nov 2000, 12:58 p.m.,
in response to message #3 by Mike Davis

> I wonder if this could also be the source of the display flashing

Yes, very likely so, if the calculator otherwise works correctly.

Viktor

      
Re: Display Power Up on a Classic
Message #8 Posted by M Currie on 3 Nov 2000, 1:26 a.m.,
in response to message #1 by Mike Davis

I have two 35's. One always flashes all the digits very briefly (a second or less) when it powers up, and the other generally does not when powered up after resting more than a minute or two, but if cycled, it flashes. One is a type 2, the other (flasher) a type 3, though they both have the same date code. They both work well.

      
Re: Display Power Up on a Classic
Message #9 Posted by Peter on 3 Nov 2000, 9:40 a.m.,
in response to message #1 by Mike Davis

My HP-35 shows just a 0 , not 0.00

Peter --

      
Re: Display Power Up on a Classic
Message #10 Posted by Tom (UK) on 6 Nov 2000, 12:47 p.m.,
in response to message #1 by Mike Davis

My '67 displays '0.00' 99 times out of 100. However in the 1/100 case it displays '-0.66_____---' (where '_' is a blank digit). It almost looks like it's in program mode (but the switch is in RUN). By pressing CLx all is OK.

My strange display also seems connected with the ON/OFF switch not comming on cleanly.

Tom.

            
Re: I guess the whole point of this question is...
Message #11 Posted by Mike Davis on 7 Nov 2000, 10:36 a.m.,
in response to message #10 by Tom (UK)

I guess the whole point of this question is, does this make a calculator not worth having as a collector item? Or, is this phenomena no big deal, for the most part?

Does a calculator that comes up immediately with "0.00" rather than flashing the digits before stabalizing at "0.00" have more value?

                  
Re: I guess the whole point of this question is...
Message #12 Posted by Tom (UK) on 8 Nov 2000, 1:15 p.m.,
in response to message #11 by Mike Davis

With my HP67 I think the non '0.00' display appears because of switch bounce on power up. What the calculator's CPU does when the power goes on-off-on in a fraction of a second will be fairly random.

I guess what is happening is that the CPU gets confused and starts executing code at a random place until that routeen finishes. Depending on what's in the CPU's scratch registers at that time and the point in the code that the CPU starts to execute will dictate what appears in the display. In my case this is constant figures that are cleared with 'Clx', other calc's may look as though they are executing a program.

As the HP67/65 has a fair ammount of program memory, I wonder if this initialy powers up with a random sequence? Does anyone know what the HP65/67 does on power-up? i.e. does the CPU go through each program location and clear it or is the design of the memory such that the memory always powers up with 'R/S' commands???

More modern designs will have a power on reset that makes the CPU do nothing until the power stabalises. I guess the older classic models either don't have this or the switches get so 'noisey' on power-up that it can't cope.

The later calc's with constant memory have the voltage supplied to the CPU all the time so there is near zero risk of it powering up with a non '0.00' display - even if the switch bounces a great deal on power up.

Each calculator will have different parasitic (i.e. stray) capacitance on the circuit board and within each CPU (silicon devices - especially older types - can vary a great deal) and this may control what happens on power-up. i.e. this is why every model of the same calc may do different things on power-up.

I don't think this should change the value of the calculator as they will work perfectly after a very short power up time. However it may be an early indication that the power switch's silicone grease is getting a bit thin. In fact this behavior makes the classic series of calculators even more endearing to me as they seem to have much more of a personality than the more modern designs. Please note I am NOT a collector, just a avid user, I use my HP67 the most even though I have: HP32/48/49.

Tom (UK)

                        
Re: I guess the whole point of this question is...
Message #13 Posted by Viktor Toth on 8 Nov 2000, 10:27 p.m.,
in response to message #12 by Tom (UK)

> modern designs will have a power on reset

So does the HP-67... or, I must correct myself since I never actually saw, or traced, the HP-67's circuit, what I know is that the HP-91 (a close relative) does have a 'power good' circuit.

> later calc's with constant memory have the voltage supplied to the CPU

Actually, that is not necessarily the case; if the C-MOS RAM is in a separate package, it is not necessary to keep the CPU powered. I believe that the CPU is in fact NOT powered in the C-MOS Woodstocks, and possibly other C-MOS HPs as well. (The cautious 'I believe' reflects the fact that although I did trace the part of the circuit that keeps the RAM chip powered in an HP-25C, I never actually double checked whether other components are completely powered down or not. One thing is sure; the switching power supply is off, so the CPU certainly doesn't receive its normal operating voltages.)

Viktor


[ Return to Index | Top of Index ]

Go back to the main exhibit hall