Post Reply 
hp-91 transistor substitution
09-20-2014, 05:10 PM (This post was last modified: 09-20-2014 05:11 PM by joseaem.)
Post: #1
hp-91 transistor substitution
    Hi all.
I'm owner of an old HP-91 calculator (year '79 more less). It makes all functions so well, but it have a 'little' problem at display. All results and all digits have the minus sign activated or bright.
Open the chassis, I discover a transistor phisically break (see picture attached).
I need a valid commercial reference of this transistor, and perhaps the oher 2 additional transistors near, because for me it's impossible find them in tables and internet.

The numbers are 'M712 3-393', 'M712 3-374', '4-071 M741'

Thanks in advance
Find all posts by this user
Quote this message in a reply
09-20-2014, 07:17 PM (This post was last modified: 09-20-2014 07:50 PM by jebem.)
Post: #2
RE: hp-91 transistor substitution
(09-20-2014 05:10 PM)joseaem Wrote:  Hi all.
I'm owner of an old HP-91 calculator (year '79 more less). It makes all functions so well, but it have a 'little' problem at display. All results and all digits have the minus sign activated or bright.
Open the chassis, I discover a transistor phisically break (see picture attached).
I need a valid commercial reference of this transistor, and perhaps the oher 2 additional transistors near, because for me it's impossible find them in tables and internet.

The numbers are 'M712 3-393', 'M712 3-374', '4-071 M741'

Thanks in advance

Hola,

The 3 transistors are from Motorola (the "M" indicates the manufacturer, and the 3 digits are the manufacturing date).

So the references are:
"3-393" --> this probably is a PNP transistor
"3-374" --> this probably is a PNP transistor
"4-071" --> this probably is a NPN transistor

Edited to add this:
My cross reference comes from this information here.
(just for reference, Motorola identification code is "185")

You are lucky because the transistors cases have the Emiter (E), Collector (C) and Base (B) pins well indicated, so that is a good starting point: it means they are bipolar silicon NPN and/or PNP semiconductors.

Take a digital multimeter set to measure Volts, put the negative probe to the Emiter pin and the positive probe to the Collector of each transistor.
If you are lucky, a voltage will be there during the normal operation of the calculator.
If the reading is positive, the transistor should be a NPN type.
If the reading is negative, the transistor should be a PNP type.

I would use fast switching high gain transistor for both NPN and PNP positions, capable of about 0,5 to 1 Amp of current, working with at least 50 Volt Vcb, and a minimum of 0,5 Watt of dissipation power.
There are thousands of possible replacements that will work on that position.

For example:
NPN: PN2222A
NPN: PN2907

One final note:
You may like to acquire the Manuals on CD/DVD from this MoHPC.
They have lots of manuals including Service Manuals as well (the HP-97 is included and it is close enough for this task).

Jose Mesquita
RadioMuseum.org member

Find all posts by this user
Quote this message in a reply
09-20-2014, 10:20 PM
Post: #3
RE: hp-91 transistor substitution
(09-20-2014 05:10 PM)joseaem Wrote:  Hi all.
I'm owner of an old HP-91 calculator (year '79 more less). It makes all functions so well, but it have a 'little' problem at display. All results and all digits have the minus sign activated or bright.
Open the chassis, I discover a transistor phisically break (see picture attached).
I need a valid commercial reference of this transistor, and perhaps the oher 2 additional transistors near, because for me it's impossible find them in tables and internet.

The numbers are 'M712 3-393', 'M712 3-374', '4-071 M741'

Thanks in advance
From your picture that does not look like a break in the transistor, that looks more like there is a coating of some kind on the back. In any case there are no transistors in the drive circuit for the display. The segments are turned on by ROM 0 through the anode driver chips. The fact that the this segment is on for all digit positions would suggest that the driver is on all the time either due to a fault in the driver chip or in ROM 0. If the anode driver had failed I would expect the segment to come on as soon as you turned on the calculator and remain on, however this symptom could also be caused by a failed ROM 0. You could isolate the anode driver by swapping them if the anode driver has failed it should cause a different segment to be lit up all the time. If it is a segment driver you may be able to get a replacement as I believe this chip 1858-0044 was used in all of the topcats, however the ROM would be unique to the 91.

The 1976 vintage 91 I just finished rebuilding, (for sale) does not even have any transistors on the the display board so it would appear that by the time the one you have was built the display board was very similar to the 97. Since this would appear to be similar to the 97s display circuit and comparing part numbers that are visible these transistors control voltage supplied to the anode drivers and if one was failed I would not expect the symptom you are observing.

The HP part numbers for the transistors are 3-393 = 1853-0393, 3-374 = 1853-0374, and 4-071 = 1854-0071 which in at least one cross reference says the industry number is CP4071 which does not seem to be helpful. I have had difficulty in the past finding substitutes for transistors used in topcats.
Find all posts by this user
Quote this message in a reply
09-21-2014, 04:10 PM (This post was last modified: 09-21-2014 05:14 PM by joseaem.)
Post: #4
RE: hp-91 transistor substitution
(09-20-2014 07:17 PM)jebem Wrote:  
(09-20-2014 05:10 PM)joseaem Wrote:  Hi all.
I'm owner of an old HP-91 calculator (year '79 more less). It makes all functions so well, but it have a 'little' problem at display. All results and all digits have the minus sign activated or bright.
Open the chassis, I discover a transistor phisically break (see picture attached).
I need a valid commercial reference of this transistor, and perhaps the oher 2 additional transistors near, because for me it's impossible find them in tables and internet.

The numbers are 'M712 3-393', 'M712 3-374', '4-071 M741'

Thanks in advance


The 3 transistors are from Motorola (the "M" indicates the manufacturer, and the 3 digits are the manufacturing date).

So the references are:
"3-393" --> this probably is a PNP transistor
"3-374" --> this probably is a PNP transistor
"4-071" --> this probably is a NPN transistor

Edited to add this:
My cross reference comes from this information here.
(just for reference, Motorola identification code is "185")

You are lucky because the transistors cases have the Emiter (E), Collector (C) and Base (B) pins well indicated, so that is a good starting point: it means they are bipolar silicon NPN and/or PNP semiconductors.

Take a digital multimeter set to measure Volts, put the negative probe to the Emiter pin and the positive probe to the Collector of each transistor.
If you are lucky, a voltage will be there during the normal operation of the calculator.
If the reading is positive, the transistor should be a NPN type.
If the reading is negative, the transistor should be a PNP type.

I would use fast switching high gain transistor for both NPN and PNP positions, capable of about 0,5 to 1 Amp of current, working with at least 50 Volt Vcb, and a minimum of 0,5 Watt of dissipation power.
There are thousands of possible replacements that will work on that position.

For example:
NPN: PN2222A
NPN: PN2907

One final note:
You may like to acquire the Manuals on CD/DVD from this MoHPC.
They have lots of manuals including Service Manuals as well (the HP-97 is included and it is close enough for this task).

Olá, como vai você ?, How are you?
Thank you for your early response. I'll try to test the items you comments. If you see the response of our collaborator in this thread (thanks a lot Paul Berger!!) affortunately this transistor not affect the seven segment display (SSD), it seems something related to powering, that is, in this case the transistor/s perhaps are additional problems ...

Along the next days I'll perform some probes and I'll tell you the results

Best regards
Find all posts by this user
Quote this message in a reply
09-21-2014, 05:12 PM
Post: #5
RE: hp-91 transistor substitution
(09-20-2014 10:20 PM)Paul Berger (Canada) Wrote:  
(09-20-2014 05:10 PM)joseaem Wrote:  Hi all.
I'm owner of an old HP-91 calculator (year '79 more less). It makes all functions so well, but it have a 'little' problem at display. All results and all digits have the minus sign activated or bright.
Open the chassis, I discover a transistor phisically break (see picture attached).
I need a valid commercial reference of this transistor, and perhaps the oher 2 additional transistors near, because for me it's impossible find them in tables and internet.

The numbers are 'M712 3-393', 'M712 3-374', '4-071 M741'

Thanks in advance
From your picture that does not look like a break in the transistor, that looks more like there is a coating of some kind on the back. In any case there are no transistors in the drive circuit for the display. The segments are turned on by ROM 0 through the anode driver chips. The fact that the this segment is on for all digit positions would suggest that the driver is on all the time either due to a fault in the driver chip or in ROM 0. If the anode driver had failed I would expect the segment to come on as soon as you turned on the calculator and remain on, however this symptom could also be caused by a failed ROM 0. You could isolate the anode driver by swapping them if the anode driver has failed it should cause a different segment to be lit up all the time. If it is a segment driver you may be able to get a replacement as I believe this chip 1858-0044 was used in all of the topcats, however the ROM would be unique to the 91.

The 1976 vintage 91 I just finished rebuilding, (for sale) does not even have any transistors on the the display board so it would appear that by the time the one you have was built the display board was very similar to the 97. Since this would appear to be similar to the 97s display circuit and comparing part numbers that are visible these transistors control voltage supplied to the anode drivers and if one was failed I would not expect the symptom you are observing.

The HP part numbers for the transistors are 3-393 = 1853-0393, 3-374 = 1853-0374, and 4-071 = 1854-0071 which in at least one cross reference says the industry number is CP4071 which does not seem to be helpful. I have had difficulty in the past finding substitutes for transistors used in topcats.

Hi Paul
Many thanks for your response.
As you indicated, the SSD was drived for two IC 1858-0044, near the display (see picture), this are a special transistor array (4x) as describe in http://www.rskey.org/CMS/index.php/exhibit-hall/130.
Next days I'll explore the behaviour of IC's and transistors as I related to JEBEM.

I'll say to you soon.

Best Regards


Attached File(s) Thumbnail(s)
   
Find all posts by this user
Quote this message in a reply
09-21-2014, 11:32 PM
Post: #6
RE: hp-91 transistor substitution
(09-20-2014 10:20 PM)Paul Berger (Canada) Wrote:  From your picture that does not look like a break in the transistor, that looks more like there is a coating of some kind on the back.

To my eyes it looks like a catastrophic failure of the transistor caused the package to split apart. I've seen that happen with over voltage or other over heating. You can find Youtube videos of people killing various components that way. As others have suggested, this may indicate there is another problem that may aslo need to be resolved.
Find all posts by this user
Quote this message in a reply
09-22-2014, 01:55 AM
Post: #7
RE: hp-91 transistor substitution
(09-21-2014 11:32 PM)cruff Wrote:  
(09-20-2014 10:20 PM)Paul Berger (Canada) Wrote:  From your picture that does not look like a break in the transistor, that looks more like there is a coating of some kind on the back.

To my eyes it looks like a catastrophic failure of the transistor caused the package to split apart. I've seen that happen with over voltage or other over heating. You can find Youtube videos of people killing various components that way. As others have suggested, this may indicate there is another problem that may aslo need to be resolved.

Yeah it is a lot clearer in the second picture he posted not so obvious in the first.
Find all posts by this user
Quote this message in a reply
09-22-2014, 06:36 PM
Post: #8
RE: hp-91 transistor substitution
LOCATED THE PROBLEM!!!
Main item, display the minus sign...
Due a transistor shorcuted in U3 anode drivers, tomorrow I will replace with a 2n2222 transistor.
Second item, the transistor break: The calculator is functional, because is a secondary energy element, making resitance lower at power demand increasing.
All thanks to HP 97 service manual.

Thanks for your interest, Jebem, Paul And Cruff


Attached File(s) Thumbnail(s)
           
Find all posts by this user
Quote this message in a reply
09-22-2014, 07:51 PM
Post: #9
RE: hp-91 transistor substitution
(09-22-2014 06:36 PM)joseaem Wrote:  LOCATED THE PROBLEM!!!
Main item, display the minus sign...
Due a transistor shorcuted in U3 anode drivers, tomorrow I will replace with a 2n2222 transistor.
Second item, the transistor break: The calculator is functional, because is a secondary energy element, making resitance lower at power demand increasing.
All thanks to HP 97 service manual.

Thanks for your interest, Jebem, Paul And Cruff

Excellent! Thank you for sharing your work.
Un Saludo,
Jose

Jose Mesquita
RadioMuseum.org member

Find all posts by this user
Quote this message in a reply
Post Reply 




User(s) browsing this thread: 1 Guest(s)