OT: Texas Instruments SR 52 Programmable Complete! A great opportunity.



#21

Ebay Link

Edited: 8 Dec 2007, 9:15 a.m. after one or more responses were posted


#22

Now that seems crazy to me. They never work!


#23

Hi,

I have already purchased items from this seller with no problem at all: descriptions always accurate and very nice items.

These machines can work if you can rebuild the huge roller of the card reader:


This can be done with a tap joint such as the one used to rebuild the 9815A capstan roller.

Another condition is that the card detector still works. I could never find any replacement for this one.

Cheers

Etienne

Edited: 8 Dec 2007, 10:43 a.m.


#24

But don't these suffer from TI-itis? You know, the keyboard failure where you press a 9 and get a string of them? This happened to my father's SR something (non-programmable but full scientific) and when I got to design school, my mentor, independently, mentioned the term "TI-itis" for exactly the same phenomenon.

As far as I know, there is no cure for the disease-- or is there?

Edited: 8 Dec 2007, 9:53 p.m.


#25

The cure is to buy an HP calculator!!!!


|
| ooo
| o o \
| ooo |
| |
| ============== |
| ooo |
| o o |
| ooo /
|
#26

None of my SR-52s suffers from the keyboard bounce problem.

But my TI-58C certainly does as well as the MBA & 59. Unfortunately, I do not know any repair method for this.

I fully agree that this simple design is less reliable than the Hp design (if I exclude the spice Series).

But the TI-58C lower price allowed a budget-challenged student to enter the marvelous world of computers.

This machine is still with me.

I love Hp products and craftmanship (up to but excluding the Hp-41C).

But it is a TI that started it all.

Cheers from France

Etienne

Edited: 9 Dec 2007, 1:53 p.m.


#27

Hi Etienne:

What you say all makes sense. What I didn't understand was how the price for TI machines could be up over $100 knowing that they are unlikely to be working...but if the SR-52 series is more solid then it makes sense.

Or put it this way: a 15C that works is worth $150 or more. A non-working one: maybe $20.

#28

The reason for the keyboard bounce was a combination of marginal debounce circuitry and a lousy mechanical design for the keys. The keys would turn soft after a few years of usage and the debounce circuits wouldn't reject the bouncing detected. This came from an
engineer I worked with in the mid 80s. He previously worked at TI and used to work in the calculator products group.

I'm curious Etienne: why do you say the 41C wasn't of good quality?

Edited: 10 Dec 2007, 10:26 a.m.


#29

Quote:
I'm curious Etienne: why do you say the 41C wasn't of good quality?

The designs of the HP-35, HP-67 and other earlier machines included battery compartments which were essentially "bathtubs". There was not an easy path by which cell leakage from the battery pack could attack the calculator circuitry. In the HP-41 design there is an open space between the battery compartment and the calculator circuitry next to the display. This provides an easy route for cell leakage to attack the calculator circuitry. The problem is exacerbated by the open end of the cell holder being placed adjacent to the "hole" in the battery compartment.

The somewhat earlier "Spice" mschine design also deviated from the "bathtub" concept for the battery compartment.

I don't understand why HP should have abandoned the "bathtub" design on later machines. I do know that TI did the same thing. I seem to remember that there was a time when the NiCad manufacturers touted their cells as "leak-free". Did the designers at HP and TI decide to believe them?


#30

Key-debouncing is easily done in software. I've done it many times in my own commercial designs. I don't know why any of these calculators should have a problem, but they do. Perhaps the delays used are inadequate for when the key condition deteriorates past a certain point. My HP-41 has a little trouble on just a few of the keys I don't use much, but my TI-59 which has NiCads is much worse. My 41 has had alkalines in it continuously for the last 21 years, and they have never leaked. In fact, I have never seen an alkaline cell leak when it still had at least 1.2V on it under load, which is the voltage at which the 41's low-battery indicator comes on.


#31

Don't forget that when the TI's were designed, maybe they didn't anticipate the key bounce issue or didn't know how to design around it?

I know that TI did eventually admit the problem in a way. Around 1984 TI was touting their new calc line and was offering a free trade-in for any old TI-30 for the newer TI-30 in our student bookstore. The pile of TI-30s the TI reps collected for the free trade was huge. I know my roommate sure was happy to not having a calc taking 8th roots inadvertently anymore.

As for the 41C, I never had any leaking battery issues either (but I can see how it can happen).

#32

Quote:
Key-debouncing is easily done in software.

Key debouncing is easily done in software (or hardware) for any reasonable switch design. The problem occurs when really bad switch designs are used, as they can bounce so long that any attempt at debouncing them effectively will also prevent deliberate consecutive keypresses from being recognized.

Some of the vintage TI calculators had switches that exhibited this problem when new; more of them were OK new but developed the problem within months of use.

In more recent years, this problem has been seen on some calculators with membrane switches. They can be designed for minimal bounce, but often are not, because for many applications it doesn't matter.

#33

If you keep your calculator face-down, you are taking a bigger risk.

However the gases are enough to screw up the electronics. I never had a 41 until I became a collector. I must say, I find it to be a fragile piece of polystyrene with breakable connections between front and back, compared to the voyagers. Once you get that gas in there from corroded batteries, you just cannot seem to stop the corrosion--or at least I can't. After cleaning everything up, it is only a matter of weeks before it is all green again and not working anymore.


Edited: 11 Dec 2007, 10:30 a.m.


#34

It's easy to criticize... but most design is nothing but a series of compromises. The 41 needed volume for the expansion roms. It needed high battery capacity for card readers and wands. You got what you got through a process of adding features to the base spec that yielded one of the best keyboard and display designs ever used. Sure, the Voyagers are better mechanically but it still has the same keyboard and display technology used in the 41. It's a 41 without the expansion. Why single out the 41? Don't like it? Use something else. Why don't we discuss the merits of the 28S battery door? Or it's case design that is guaranteed to fail with use.

I'll add that most of the warts in the HP calculator line are mechanical in nature. The only really ugly electrical design flaw that comes to mind is the battery charger circuit the 20 series.

Just how many really *PERFECT* designs have you ever encountered in your lifetime?

A Western Electric black dial phone comes to mind...

PS: Just change your batteries in your 41 once a year, never mix brands or age cells and you'll never have a problem. Same goes for the 48 series too.


Edited: 11 Dec 2007, 11:04 a.m.


#35

Hi Randy:

Yes you are correct about relative merits and compromise and effectively overall quality.

I guess I should have spoken further about the relative fragility. Clearly, the 41 as designed produced more than adequate service life. These downsides are relative downsides 20 years out of production.


#36

I think relative fragility is great way to express things. After all, ten years was probably the projected life span when designed. Now that we're pushing thirty, many things get pointed out as flaws when in fact the engineering back then was pretty darn sound.

IMO, it's all a matter of perception. How many times have you ever said to yourself, if I only knew then what I know now :)

#37

Quote:
However the gases are enough to screw up the electronics. ... Once you get that gas in there from corroded batteries, you just cannot seem to stop the corrosion--or at least I can't. After cleaning everything up, it is only a matter of weeks before it is all green again and not working anymore.

Use vinegar!

Like a crusty car battery, cleaning is insufficient. You have to neutralize the traces of electrolyte which remain in the pores, cracks and crevices or else the crusty corrosion quickly returns. Unlike a car battery which uses acidic electrolyte, an "alkaline" battery has, you guessed it, alkaline electrolyte.

Clean all the crust and chunks off that you safely can, and use vinegar to neutralize what is left. Sometimes then you find you can clean a bit more, so do that and use the vinegar again.

I like to finish off with distilled water, several wipes or a thorough flush, in order to remove all the vinegar. Stopping alkaline corrosion only to induce acidic corrosion isn't nice. :)

sdb

#38

His name is Perry,

Perry W. Kaminsky.

Please read TI's patent application US3973091 filed in 1975.

He is the good guy! This design is smart and worked perfect with the good old SR-50, SR-51...SR-56 Programmable.

(A good reader is availabe from ESPACENET:
http://v3.espacenet.com/origdocDB=EPODOC&IDX=US3973091&F=0&QPN=US3973091 )

The original design used gold on gold, later the gold was getting thinner and thinner and finally we had silver on silver designs.
If you open the "worst design", e.g. a TI-55 II you'll notice badly corroded wires under the adhesive foil.

Regards,
Joerg

#39

Hello,

My main issue with the Hp-41C design is the zebra connector used to connect the mother board to the keyboard/display assembly (on the fullnut models):

Correct connection is ensured by sufficient pressure on this connector. This pressure comes from the plastic screw posts securing the 2 half cases.

My view is that the design of the connector on earlier calculators (with golden pins going through holes) is more reliable on the long run.

See this connector on the bottom of one of my Hp-80:

I enjoy using the Hp-41C series very much, however, even when Hp solved this connector issue with the halfnut evolution, the Hp-41C keyboard is still very difficult to repair compared to the Classics series keyboard.

Cheers from France.

Etienne



Edited: 12 Dec 2007, 2:49 a.m.


#40

Quote:
even when Hp solved this connector issue with the halfnut evolution, the Hp-41C keyboard is still very difficult to repair compared to the Classics series keyboard.

The 41C family both original and halfnut, were designed such that the keyboard repair was very easy. Remove top case, discard, install new top case. It's right in the service manual.

Oh, did you mean that you wanted it to be easy for you to repair? Sorry, that wasn't in the design spec. In fact, it's generally not in the design spec for any consumer electronic product. Consequently, I don't think it's reasonable to criticize the design on that point.

I agree with you that the contacts used in earlier HPs had better long term reliability. They also cost more, both for the contacts and the gold plating of the PCB. The zebra strip was obviously intended to meet the product reliability objective at a lower cost. As far as I know, it did so.

I suspect that the reason for the better contacts in the Classic and Woodstock series was that making consumer products was new to HP at the time, and they didn't have a good feel for how much overengineering was necessary to meet the reliability objective for a product that might be subjected to a fair amount of physical abuse. I don't think it was specifically intended that the Classic series last for more than 30 years, even though they've generally done so.

In the Spice series, they went too far the other way with the press-fit assembly, and had a high defect rate eventually resulting in a redesign.

Edited: 12 Dec 2007, 5:22 a.m.


Possibly Related Threads...
Thread Author Replies Views Last Post
  Converting Great Circle Navigation from 41C to 42S Bill Triplett 11 958 11-13-2013, 07:24 AM
Last Post: Kimberly Thompson
  prime programming / great circle formulae Geoff Quickfall 15 933 09-27-2013, 07:16 PM
Last Post: Kimberly Thompson
  SandMatrix Released - The trilogy is complete. Ángel Martin 3 474 08-30-2013, 05:35 PM
Last Post: Marcel Samek
  who uses programmable calculators? John Ioannidis 24 1,694 06-18-2013, 04:11 PM
Last Post: Steve Simpkin
  [alpha]SR wp34s Andrew Nikitin 9 774 06-17-2013, 11:48 PM
Last Post: Andrew Nikitin
  Great news - Vicinno's HP 15C Scientific Calculator iPhone app is FREE now John 21 1,668 06-07-2013, 05:49 AM
Last Post: Mike (Stgt)
  The 'Complete' Collection Keith Midson 8 1,020 05-12-2013, 09:52 PM
Last Post: Namir
  HP85 Programmable ROM cardtridge 82929A-service ROM not working- inaki 2 428 04-25-2013, 08:08 AM
Last Post: inaki
  Rather large missed opportunity - a pile of HP71B's Chris Smith 3 422 04-19-2013, 07:30 AM
Last Post: Chris Smith
  My Mostek collection is complete Michael de Estrada 15 1,558 04-10-2013, 10:01 PM
Last Post: Michael de Estrada

Forum Jump: