HP Forums

Full Version: Which side of the glass case do the Hp's belong on?
You're currently viewing a stripped down version of our content. View the full version with proper formatting.

Having lurked for some time, I have grown curious about how everyone treats the calculators they collect.

Do you keep them under glass, carefully locked in some curio cabinet, lest their aging keys suffer wear from dirty fingers?

Or do you adhere more closely to the TOY STORY philosophy that a toy becomes miserable just sitting on a shelf and must be active?

I myself teach everyday with one of my HP sidekicks wheezing through the smog of chalk right beside me. Perhaps that kind of abuse is sacrilegious to a few of you serious collectors. Somedays I'll even take along the Novus, lest it get bored.

What say you? What's the more proper fate for your noble HP's?

Hi Richard.

My HP41CV (early fullnut Sliverbird) is a bit mechanically fragile
at the moment so I am handling it carefully until I get time to do some structural repairs on it. But it still runs perfectly and is fully functional after 24 years and five repairs. It gets use doing impedance calculations on capacitors and on timing values for LM555 oscillator circuits which I use for clocking some digital projects I am doing. It has a TIMER module also and so is used to alert me to closing Ebay auctions and I wrote a very simple PROJECT MANAGEMENT program with it to help goad me into putting in the neccesary time to get my longer term projects out the door on time.

If you and others are curious you may like to look at my website:


I have a few photos there of some of my digital electronics projects. The latest one centers on a tool I am builidng (a five channel serial Logic State Analyzer) to assist me in designing and building hardware add-ons for the HP41C series machines among other things...

I think any self-respecting HP engineering calculator would rather
"die with it's boots on" in a chalk saturated classroom helping the engineers of tommorrow or end up MIA on a construction site rather than quietly languish in a glass case somewhere. But collectors are
also an important part of the "user mix" that constitutes this community of people within the forum. You find a lot of helpful people here and I am delighted to see what others are doing and offer help where I can. The speed and quality of responses to my own questions makes me feel really confident about "hanging out"
on the forum and I have met some VERY INTERESTING people here.
Some are rather qualified and experienced. All are pleasant and helpful to each other. If RL was this way there would probably be no war (just a few healthy debates from time to time ;-)

Welcome aboard!



I keep most of my HPs on a shelf, some behind glass, in a display case in my dining room, the rest in the basement. However, I don't hesitate to pull one out and turn it on when someone comes over and shows some interest. (besides "Calculator collecting? Why?") Sometimes some of my machines travel to displays in local libraries or schools. (Of course, there they're behind glass in a safe environment.)

I also have a few I keep out to use: a beat up HP 15C next to my computer for miscellaneous calulations, an HP 11C (in pretty good shape) in my desk at work for occasional use, and an HP 33S on my desk at home for balancing my check book.

My HP 16 sometimes comes out when I'm working with number theory.


Bottom Line

My 41cx is going on 20 years of perfect daily operation with no repairs. There was a time however that it was getting operated in a dusty environment, so I kept it in a clear zip-lock bag, with the HPIL cables (to the IEEE-488 interface converter) coming out a corner and the extra bag plastic folded behind and taped. So in a sense, you could say it was getting heavy daily use "behind the glass."

Hello Don,

I just had a look at your really impressive documentation for repairing a 41. What I missed is a closer look at the keyboard-PCB. The "8"-key of my 41C has the problem that you nearly always have to push it twice before the 8 is accepted. Did you had such a keyboard problem once, too ? And could you show me how to repair it ? Thank you for your answer in advance.



See here two rare pics of opened display cabinets. Typically I touch those items only with cotton gloves. Only a real HP200LX I have still in use, all others I use as emulation only.


Hi Karl.

I did have a serious failure in the keyboard of my 41CV in 1994.

The actuators for the keys are moulded in the keys themselves.
I found that the ALPHA and USER plastic spikes broke off.

Others have said the keyboard is not repairable.
I would say this is partly true and if you have no great experience with fine soldering on CMOS circuitry your chances aren't that good at all. Experiened people will find it a tedious exercise which involves a total strip down of the calculator. My repair took me about five to six hours. But I don't rush repairs like this.

The most delicate part of the repair is to unsolder the LCD module from the keyboard PCB. Once that's done one can cut away the many plastic heat stakes holding the keyboard onto the front panel to access the key mouldings themselves to repair them.

At this point you have to determine if it's a failed key or the
actuator has broken away. This is readily apparent. The key moulding has to be repaired and then the keyboard PCB has to be re-staked to the front panel. The main problem is repairing the key acuator so that the keyboard will function properly with a pretty long life. Due to the limited length of the heat stakes, I believe keyboard repair is a once-only procedure.

Once that is done, re-assembly proceeds with components going back where they came from with due care to avoid static damage.

From what I saw of the keyswitches themselves I believe spray solvent or immersion / ultrasonic cleaning would not work as they are pretty well sealed.

Hope this helps. Happy to repair your 41 if you want.
If so, please write to me.

Don Wallace

Hi, Richard:

I own more than a hundred vintage calculators, mostly HP and
SHARP BASIC handhelds, together with their manuals, accessories, etc.

But for three, all of them are safely stored within an enclosed, humidity-controlled cabinet in a relatively cold place, each particular calculator inside a hermetically thermo-sealed plastic bag. This keeps interaction with light, heat, air, and humidity to a minimum. Before storage, each calculator has been carefully cleaned up, both inside and outside, and of course with no batteries or power cells in them. This guarantees as much as possible that except for time itself, most aging and degrading agents are left out.

As for the three models excepted from this treatment, they are an HP-15C, a full HP-71B, and a SHARP PC-1350. All of them are in very good condition (the 71B being actually mint), and see frequent use. The HP-15C is further protected by being tightly enclosed in a very clear plastic bag, which prevents any physical contact with its surfaces, thus avoiding stains, fingerprints, dust under the display, moisture, etc., while allowing perfectly comfortable use.

The other two models are not so delicate and do not require such extra protective measures, they're simply treated with care.

Overkill you might think ? Not so. These measures ensure that my valuable calculator collection stays in pristine condition for years to come, so that I can take out any of them some years in the near future and have the extreme pleasure of holding in my hands a, say, pristine, not-a-single-dust-speck, not-the-slightest-blemish, shiny and new HP42S, HP-15C, or HP-41CX, say. Proceeding otherwise I would be denied this exhilariating experience.

Do I not miss having them at hand at all times, using them, etc ? Well, I *do* have an HP-15C, full HP-71B and SHARP PC-1350 available at all times, and they do get used. As for the rest, I did make extremely good quality digital pictures of all of them, and can look at them at will.

And re using them, with such fantastic emulators as V41, Emu42, Emu71, Nonpareil, and SHARP PocketEmu, to name a few, my vintage programming needs are more that fulfilled, and with better facilities to boot (keyboard, mass storage for calculators that had none whatsoever, etc).

Best regards from V.

It would be worth trying to clean the contacts without dismantling the keyboard.

Take the machine apart (remove the back case, the logic board, and the black 'paper' cover over the keyboard PCB). Now get some fine strands of wire (e.g. from a piece of multi-strand wire) and grip them in a pin chuck. Trim them all to the same length so you have a tiny wire brush. Put it through the via (hole) behind the defective key, and carefully brush the contacts. Maybe drop a _little_ propan-2-ol down there as well. Then put it all back together and try it.

If it doesn't help you've not lost much. If it does, you've saved the problems of re-assembling the heat-staked keyboard.

I've said many times that the _appearance_ of a calculator doesn't matter to me. What interests me is how it works, what it does, the electronic design, the firmware. So I use them, take them apart, probe them with a logic analyser, solder wires to the insides, and similar things

I have little interest in collecting lumps of plastic that I never turn on. Therefore my HPs get used. Well used in some cases. I always have at least 2 41's, a 67, a 71B and several IL peripherals with good batteries and 'ready to go'.

I treat most of my other things like that too. Obviously for optical stuff (cameras, etc) I make darn sure the lenses don't get damaged -- that would affect the performance. But a 'ding' in the paint wouldn't bother me at all.

Good suggestion...


Hi Richard,

I have about 20 calculators. None of them are "stored away" though I try to take good care of them.

I keep on top of the battery situation. I am quite certain that use keeps a battery from leaking--at least that appears to be the case! (My wife has a sharp elsi-mate that is 20 years old, and still has the original battery!).

I like to rotate them into use, too. It is fun to go back to an 11c after having been awy from it for a long time.

Or to switch to a 32s 50th anniversary edition from a 32sii--the color scheme is different.

I also find that now that I have a working collection, I think about programming in a different way. I go the "easiest route" and so if I know that the 48gx has a good way to handle something, I pull it out. Sometimes I have three on my desk at once--since each one is already programmed for some particular task.

For travel, I use less valuable machines: I used to take the 20s around, but now I use the 33s. I carry a 49g (not the new G+) when I want to have an RPL machine on the road.

Hi Karl,

I just rxed a hp41C, very old, totally D.O.A.

In the process of diassempling the machine, I will photograph the keyboard PCB and post my pictures for you.

It appears to me that you are very unlikely to get alcohol into the
keyswitches themselves from the back of the keyboard. There are very small "plate through" holes in the PCB, but they are too small to allow alcohol to go through, I think. Capillary action will stop it. The photos will show you what I mean...


Hi Karl,

I have quickly posted some photos of the HP 41C keyboard PCB.

No comments yet. I'll post some shortly...

Hope these help you.


Hello Don and Tony,

thank you very, very much for your suggestions and the really great photos. As the photos show the 41-keyboard has nothing to do with the simple to repair 35/65/67-keyboard anymore. Well, the "8"-key is still working although one often need 2 presses. And because I just got 41cx for my collection without this problem I'll let this as is. Many thanks for your repair-offer, Don, I may be come back to you at a later date.

Hope one day I can return the favour.

Kind regards


I have over 500 hand-helds. The great majority are in working condition. I have another 25 desktops, over 100 slide rules, and numerous mechanical calculators including two Fridens. I have carrying cases and documentation for many of the units. Putting all that material under glass or in special packaging isn't possible for me. I keep the calculators in an air-conditioned environment with battery packs removed where possible.

Some of my biggest problems are with devices where circuitry has been damaged by cell leakage and with some of the older hand-helds where the rechargeable NiCad's are hard-wired in. I clean up the residue due to cell leakage as best I can.

I have thirty HP's including three HP-11C's, four HP-12C's and four HP-41's. Only the HP-11C which is one of the early models with the glitch and the HP-41C which has gold balls and exhibits bug 2 qualify for special care.

The easiest to care for are the Sharps which are by far the best built units. When I see a Sharp in a thrift store or at a garage sale I expect that it will work properly. By far the hardest to care for are the NOVUS/National Semicondctor models. When I see one of those I am surprised if it works properly, and units which were once working properly have gone bad on the shelf. The rest of the manufacturers are somewhere in between.

Just like the last post, I have at least a dozen models from the 20S up to the 48GX (got rid of my 49G+). Some I have two of, so I don't mind getting more use with the "back-up" model (e.g., 20S, 32SII, 48G, G+, GX). I try to keep them all rotated into use and stay on top of the battery situation, too. The calculators get jealous if one gets more use than another! I take extremely good care of each of them, but know that some day I'll die, and what a regret I'll have if all I did was keep them under glass and not get any use out of them. (Just my opinion, of course.) Mine don't see chalk dust or dirty hands that always want to borrow them, and I use a cheap Casio when I go to the operating room (medical physicist) just in case something happens and a doctor "contaminates" it. So my HPs have an easy life and they "know" they are appreciated and used routinely. My current hobby is trying to make a medical physics PAC for my own personal use at work.

Hello Don,

thanks a lot for adding comments to the pictures of the 41-keyboard.

Kind regards


No problem, Karl.

You may find my pages get more interesting: I now have some 41 hardware I have been waiting on to build a Clonix ROMPACK.
I will be steadily working on my interface projects and posting photos as I go. So if you are interested by all means keep an eye on my pages...


I have quite a number of machines, all brands including HP (!) and also Sharp which are wonderful hardware and software-wise.

They are stored vertically (so keyboards are not pushed onto) in cupboards (no light, no dust) in a bedroom normally heated/ventilated. They are wrapped in foam or in the original pouch, whichever is available, and only a few have batteries. I have never experienced battery leakage.

Few machines get real usage, as I like to work on real problems and only recent machines provide semi-adequate speed. So usually I get a recent machine (HP48, a Sharp, others). Most machines are on a permanent rest, and some talk to me a few times each year.
I always fear I can damage my beloves LED models.

Fortunately everybody thinks I'm crazy, so they stay clear of the stuff. Behaving bizarrely is sufficient, and for most of us it must be easy !

As of now, no machine died on the shelf (including the Novus). Actually, he older the machine the better its quality it seems (apart from 'budget' ones like some Radofins or Novus etc).

People think you are crazy, I collect sliderules and they think I am totally bonkers!!!

So, tell us about your slide rule collection.... Have web photos yet? Mine (early 1970s) still work, and the "battery" has a lifetime warranty!

Hi Chris , John.

Hey do you have any of those circular aviation ones for doing flight plans / nav?

I always thought circular slide rules were cool, being "different".
Maybe thats why the 41CV appealed to me when it came out...


Hey do you have any of those circular aviation ones for doing flight plans / nav?

That would be the E6B. Somewhere I have a smaller, less expensive one for aviation that didn't do as much as the E6B.

Flight computers - a.k.a. "whizz wheels" - have two sides. One is a fairly conventional circular slide rule which is used for time/speed/distance problems. It often has markings for useful conversion factors (e.g. litres/gallons/lbs/kg) and of course, has a prominent index mark against 60, since most time/distance calculations done in flight involve minutes although speed is expressed in knots.

The back side is unique to navigation devices and usually has a panel which slides up and down behind a rotating perspex wheel; this is used for solving wind triangle problems (e.g. working out the heading to fly in order to maintain the correct track for a given cross-wind and what the resulting groundspeed will be). It's a very simple vector calculation, and the flight computer lays it out graphically.

A few (e.g. Jeppesen CR series) don't have the large rectangular slide, but are completely circular. They (probably) use a slightly more tricky trigonometric technique to solve wind problems, and make it harder for the user to intuit what's going on. However, the smaller ones like the CR-5 can fit in a shirt pocket.

I occasionally pull mine out and check that I can still do things the old-fashioned way. Funnily enough, I've never had to replace the batteries, either . . .


--- Les


Hi Les,

Yeah, the Jeppesen is the one I was thinking of...

About batteries... That reminds me of a joke I coined when I
was baby-sitting an SBS server for some guys who (think they?)
run a communicatins company:

Q. Whats the difference between a Notebook PC running Windows
and a Spirax notepad?

A. A spirax doesn't crash or need batteries.

Since I am the original purchaser of my HP-41CV (in 1981) I feel no problems with using it regularly, since that is why I bought it. I figure that that way I will replace the batteries regularly.

I do, however, keep away from the kids. And it does not go travelling. I have a new (vintage 2003) HP for that, and occasionally I take a "beater" 6-inch slide rule for fun. I probably am showing my age to say that I used a slide rule in high school.

As far as I can determine, they are very resistant to normal wear and tear.

I have about 170 slide rules in my collection including the K&E Log Log Duplex Decitrig that I purchased while a freshman engineering student in 1947. I also have several of the navigation slide rules.

One of the most unusual slide rules in my collection is a Faber-Castell 67/22R which has a 12.5 cm slide rule on the front and an Addiator (one of the push-down adders) on the back. I think it was an attempt to compete with the early four-bangers.

One of the problems with the circular slide rules is that the scales closer to the axis are shorter. Typically, the C and D scales are on the outside and the K scale is innermost and practically useless.

Some slide rules do deteriorate. Many yellow with age. The slides on the early metal Picketts (made with magnesium, I think)tended to corrode and bind. In the 1940's there wasn't a good solution for that. Now one can make them work as smooth as silk with a little WD-40 on the slide.


I do have circular sliderules, Otis Kings and of course straight ones. In all about 60 of them. I also have a sliderule used by the army for calculating the elevation angle for firing artillery shells!!!


Hi Palmer,

What sort of engineering did you do?
What is your most enjoyed aspect of retirement? (assuming u r ...)

What is your favourite computing device of all time?
What is your most loved branch / area of maths?
What is your most hated? [mine is D.E.s and statistics ;-P]

Just curious ;-)


Hi Dean,

Hey man, I really like the disciplinarian "sub-text"...

I don't have any kids (Thank Chr....)


The small Voyager and Pioneer pocket calculators I own, are rotated on a weekly basis (but I still like most the 32SII, and dislike the cheeper 20S, 21S etc...). They sit in a corner of my desk in their hull and wait for their job.

The 41CV, 48SX, 48GX I own are the one I hold "under glass" with all the fine accessories arranged. But even they are still in use for grater programming stuff (or data collection) from time to time.

I received my BS from the University of Minnesota in 1950. After a year of half-time grad school and half-time teaching freshmen mathematics I entered the Navy in 1951. I left the Navy in 1956 and joined Honeywell. I worked with automatic flight control systems and inertial navigation systems on aircraft such as the F-100, F-101, A-11, YF12, etc., from 1957 through 1989.

My retirement activity has been limited by multiple bouts with cancer.

My favorite computing device by far was and is the Radio Shack Model 100. That device was primarily designed for use by reporters. I found that Microsoft had provided a BASIC with astounding scientific capability. It's default mode was double precision which used 14 digits. I did some fairly exhaustive testing of the trigonometric functions. The largest errors I found were 5 in the fourteenth place. The display could be used for graphing with a 240 x 64 pixel range although I had to write my own graphing routines. This was is in 1983. Casio would not provide graphing capability until the fx-7000G with 96 x 64 pixels in1985, HP not until the HP-28 with 137 x 32 pixels in 1988 and TI not until the TI-81 with 96 x 64 pixels in 1990. The Model 100 also provided a full size keyboard which was a real advantage to me since I could touch type, something adults can't do on computers with baby sized keyboards. A major deficiency of the device was that the external power supply would not charge the batteries.

I did a lot of work with linear regression and curve fitting, not because I particularly liked it, but because my work required it.

Hi Palmer.


I had an old friend who has since passed on who was quite berzerk over two hardware items: His HP 11C and his TI-99/4A. Don Curry could rattle of a polynomial expression for any trig or trancendental you cared to ask about. Amazing guy, who used to install ground station equiment (including help with the odd dish) for Marconi in the 50's and 60's in South Africa and England, I think.

He shared your interest in accuracy of trigs and in using curve fit and polynomial stuff. He was doing a high accuracy planetary ephemeris in turbo pascal (bcd) and was, like myself, into Astronomy as an amateur. We had trouble with the int function and with cummulative rounding errors. He was very into high accuracy versions of Basic, Pascal and to an extent, Forth. In fact I got him a copy of Power Basic v 1.0 from Bob Zane;
it is quite good, in it's way. Way better than anything Microsoft had at the time.

My contact with Pascal, Forth and Mathematics has crossed paths with quite a few people who have similar interests over the years.

Clear Skies, Palmer!


I found that Microsoft had provided a BASIC with astounding scientific capability. It's default mode was double precision
which used 14 digits. I did some fairly exhaustive testing of the trigonometric functions. The largest errors I found were 5 in the
fourteenth place.

Here's how the MSX, an 8-bit computer with a Microsoft developed BASIC, behaves in this usual
trigonometry test:

10 K=ATN(1)/45
20 DEF FNASN(X)=ATN(1/SQR(1/(X*X)-1))
30 DEF FNACS(X)=ATN(SQR(1/(X*X)-1))

According to Mike Sebastian's Calculator Forensics Results, the Radio Shack TRS-80
Model 100 Computer returns 8.9999999956625, which means both interpreters may share the same
code. Very likely, since the Zilog Z80 is backward compatible with the Intel 80C85

That's Microsoft's implementation of trigonometric functions on the MSX computer:

From "The MSX Red Book":
(Avalon Software. The MSX Red Book. Pangbourne: Kuma Computers, [c.1985].)
(This book is not on sale anymore, but a text file can be found at


... The function is then computed by polynomial
approximation (2C88H) using the list of coefficients at 2DEFH.
These are the first eight terms in the Taylor series X-
(X^3/3!)+(X^5/5!)-(X^7/7!) ... with the coefficients multiplied
by successive factors of 2*PI to compensate for the initial


... The function (ATN) is computed by polynomial approximation (2C88H)
using the list of coefficients at 2E30H. These are the first
eight terms in the Taylor series X-(x^3/3)+(X^5/5)-(X^7/7) ...
with the coefficients modified slightly to telescope the


2DF0H -.69215692291809
2DF8H 3.8172886385771
2E00H -15.094499474801
2E08H 42.058689667355
2E10H -76.705859683291
2E18H 81.605249275513
2E20H -41.341702240398
2E20H -41.341702240398
2E30H 8 ATN
2E31H -.05208693904000
2E39H .07530714913480
2E41H -.09081343224705
2E49H .11110794184029
2E51H -.14285708554884
2E59H .19999999948967
2E61H -.33333333333160
2E69H 1.0000000000000


Edited: 15 Sept 2005, 12:23 p.m.

Hi Gerson,

Correct me if I am wrong but the Intel 8085 and the Zilog "Z80CPU"
are contemporary and not 100% op code compatible with each other.
They are both op code back compatible with the Intel 8080.

There was a math library (with a few really nasty bugs, most notoriously the INT function code) written in C, which got used
a lot through the 1980's... found it's way into quite a few versions of (personal computer) BASICs... [bugs - Yuck!].

Pascal we found was better, but alos had it's limitations.
I believe JPL used FORTRAN for the required precision to manage
the voyager series spaceprobe flight plans. Interesting.

Anyone heard of PAPC.exe, a variable precision HP calc simulator.
Runs user code in MS DOS... Anyone want a copy?
PAPC=Programmable Arbitrary Precision Calculator


Hello Don,

By what I can remember, the Z80 was designed to be backward compatible with the 8085, at least at software level. I had a course on the 8085 at work, back in 1986, and soon noticed the compatibility when I started to study the Assembly language of the 8-bit Z80-based microcomputer I had then (and still have :-)

As I wrote from memory, I thought the early Microsoft BASIC had been written originally for the 8085 microprocessor. In fact, it was written first in the 8080 Assembly Language and later ported to 8085 and Z80 microprocessors (and not taking fully advantage of the more complete intruction set of the latter), according to my copy of the MSX Red Book in Portuguese (it took me a while to find it).

I had a hint you'd be the first one to reply. Thanks for your remarks,


The Z-80 predated the 8085, so it can't be backward-compatible with it.

In fact, the 8085 only added a couple of new instructions over the 8080. The major new features were the use of a single 5V supply (the 8080 also required -5V and +12V), a reorganised pinout (to work with the related RAM/ROM & I/O chips) and the incorporation of some serial I/O (which accounts for the new instructions).

The Z-80, by contrast, provided an alternate register set and a lot of new instructions, though the assembler syntax was completely different. Incidentally, the Z-80 was designed by Federico Faggin, who had earlier designed the 4004 and the 8080 at Intel, but left Intel in 1974 to found Zilog.

Both were designed as a superset of the 8080 architecture, though.


--- Les


Thanks, Les, for definetely clarifying the matter, and Don for first pointing me out the mistake.
I've just found the old 8085 stuff, but that took a little longer... I had written below the 8085 syntax the Z80 correspondence when I began to study the Z80 Assembly: MVI M, byte -> LD (HL),byte; INR -> INC r; XRA A -> XOR A, etc.



Good one, Les. I didn't wanna go into painful detail on it all and you saved me the hassle.

Cheers, Man!


IIRC, the 8085 only added two instructions, RIM (Read Interrupt Mask) and SIM (Set Interrupt Mask), which could be used for software-controlled serial I/O.

IIRC, the 8085 only added two instructions, RIM (Read Interrupt Mask) and SIM (Set Interrupt Mask), which could be used for software-controlled serial I/O.

You have a very good memory, John - I racked my brains over that, and could only remember that there were two new instructions, but not what they were!


--- Les


The 8085 actually added about a dozen new instructions. They were deliberately designed in, not the accidental result of partial decode, but Intel management decided to leave them undocumented. A pity, since they were specifically designed to make compiled code more efficient.

Presumably Microsoft did not use them in any of their 8080/8085/Z80 software.


The 8085 actually added about a dozen new instructions. They were deliberately designed in, not the accidental result of partial decode, but Intel management decided to leave them undocumented. A pity, since they were specifically designed to make compiled code more efficient.

Presumably Microsoft did not use them in any of their 8080/8085/Z80 software.

Any links to this? I always thought RIM & SIM were the only new 8085 instructions; I'd heard there were one or two 'practical' hidden ones on NEC 80C85 chips.

Bill Wiese

San Jose CA USA

There were ten undocumented instructions:

  • HL := HL - BC
  • DE := HL + imm16
  • DE := SP + imm16
  • memory [DE] := HL
  • HL := memory [DE]
  • arithmetic shift right HL
  • rotate DE left through carry
  • jump if status reg bit 5 set
  • jump if status reg bit 5 not set
  • RST 40 if overflow

The first five in this list would be especially useful in compiled code.

I have no idea whether any of the CMOS variants included these instructions; the CMOS parts might have been redesigned from the published specifications.

I've seen some vague references to the 8085A having some bug fixes compared to the original 8085, but nothing definitive, and I'm dubious. It seems more likely that it just had improvements to the electrical characteristics, as a result of layout changes or process tweaks. The 8085AH was redesigned in HMOS, Intel's high-density NMOS process, and ran faster. AFAIK, the undocumented instructions existed in all of the Intel NMOS/HMOS parts.