The Museum of HP Calculators

HP Forum Archive 21

[ Return to Index | Top of Index ]

WP34S conversion, yet another but with a difference
Message #1 Posted by Geoff Quickfall on 21 June 2012, 5:10 p.m.

Hi all

yet another set of hardware conversion photos but I think these may add to the knowledge set. I will be including these and more for the book and have added two youtube videos (links at end of post) for those of you that have not yet considered the WP34S.

I have used hardware that Katie Wasserman proposed. This includes a trimming cap for the quartz cyrstal assembly which allows the timing accuracy to be adjusted. I have posted this assembly as it is different enough from Alexanders excellent PDF to add to the options.

So here are the parts:

timimg hardware:

-crystal..............32.768kHz, 6PF cylinder, -capacitor............ceramic 6pF 50V radial -trimming capacitor...2.8 - 20 pFd variable cap

printer hardware:

-emitter..............IR 3MM HI EFF 940NM -resistor.............390 Ohm 5% axial

Timing hardware:

Printing hardware:

I started by soldering the components to the PCA. I used a handheld, battery operated, fine point soldering iron. All points were pre soldered.

Timing Hardware in place: here we see the quartz crystal and ceramic cap in place. The C4 solder points are used for the trimming capacitor in the next photo.

The trimming capacitor is now added. Note that under the blue capacitor's leads, I have placed some electricians tape (small rectangle) to cover to solder points not used and to prevent shorting with the cap legs.

Printer solder points:

In addition to adding hardware to the PCA one must also consider the calculator case. In this example the back shell is a mass of strengthening baffles and ridges which cause problems when hardware is added to the PCA with respect to clearances.

In this photo the yellow boxes represent areas where I removed the ridge. The red box indicates the new hole drilled to allow access to the trimming cap for adjusting the timing accuracy. The trimming cap is also slightly proud of the back shell so the hole not only allows access but also allows the case to close without impacting the cap. You can see the placement I used for the IR emitter and resistor and how it is routed around the LCD and through the milled areas of the bottom shell to the solder points.

The wires are laid in using the milled areas so that they will not prevent the secure closing of the back shell. I used a 7/64ths bit to drill a hole in the top side of the calc. This allowed a tight friction fit of the emitter for the printer. It also allowed me to position the emitter so it is pointing parallel to the surface of the calculator.

Here is the back shell with the new access hole for the trimming cap. To position the drill, I applied a small drop of paint to the top of the trimming capacitor once it was soldered in place. I then assembled the calc and the paint on the cap was transferred to the inside back of the back shell marking the drilling point:

After placing the back shell (snapped not screwed in place) I checked for fit. Were the wires interfering with a correct fit and etc. Once happy, insert screws and tighten.

Of course I did test prior to the screws being applied. Here are the you tube results:

stopwatch youtube

printer youtube

Stop watch:

Printer output:

Special thanks to Pauli and Marcus (his mother makes good cookies). Also thanks to Katie for comming up with the added adjusting function for the stopwatch accuracy and Alexander and all the others that have contributed.

Now to program and work on chapter 13 "WP34S conversion".

Cheers, Geoff

edited:

just executed the 'WHO' command so I am adding names to Pauli and Marcus: thanks also Walter and Neil!!!!

32.768 kHz. Or 32,768 Hz. :-)

Edited: 22 June 2012, 11:37 a.m. after one or more responses were posted

      
Re: WP34S conversion, yet another but with a difference
Message #2 Posted by Gerson W. Barbosa on 21 June 2012, 6:17 p.m.,
in response to message #1 by Geoff Quickfall

Nice pictures and detailed instructions!
I have used thin wires from a 30-wire IDE cable, but I think a higher AWG flat cable would be better.


This is Harald's board (USB/IR version)

Regards,

Gerson.

            
Re: WP34S conversion, yet another but with a difference
Message #3 Posted by geoff quickfall on 21 June 2012, 6:24 p.m.,
in response to message #2 by Gerson W. Barbosa

The board is a nice option!

The thinnest wires I had in supply were from a standard telephone system with the outer shell stripped.

Still waiting for the overlay but getting used to key positions so far.

Cheers, Geoff

                  
Re: WP34S conversion, yet another but with a difference
Message #4 Posted by Walter B on 22 June 2012, 6:14 a.m.,
in response to message #3 by geoff quickfall

Please see (and use) p. 126 for the time being :-)

            
Re: WP34S conversion, yet another but with a difference
Message #5 Posted by Paul Berger (Canada) on 21 June 2012, 9:16 p.m.,
in response to message #2 by Gerson W. Barbosa

I did mine using 30 gauge wirewrap wire most connections and used a slightly heavier gauge of wire for the Batt and ground connections. I did not glue down my board inside but used a plastic sleeve that fits snugly over the shaved down post and cut off just slightly longer that the post, with the back cover in place it hold the board firmly in place.

Pictures

                  
Re: WP34S conversion, yet another but with a difference
Message #6 Posted by Gerson W. Barbosa on 21 June 2012, 10:35 p.m.,
in response to message #5 by Paul Berger (Canada)

I like your sleeve solution. I melted a small portion of the plastic post next to the board just enough to keep it from sliding when the micro-USB cable is inserted. The pressure of the plastic case against the USB connector keeps the board firmly in place. I think next time I get one ready made :-)

            
Re: WP34S conversion, yet another but with a difference
Message #7 Posted by Harald on 22 June 2012, 5:54 a.m.,
in response to message #2 by Gerson W. Barbosa

Hi Gerson,

thanks for posting the pictures. But I noticed there is one thing you forgot to do. Component LB2 is still in place. You should remove that, as it now connects the battery to the usb supply. Which means you are charging the CR2032s to 3.3V when the calculator is connected to the USB. And since they are not rechargable batteries, this is not a good thing. Might lead to them leaking. After doing this, you will be able to check if everything works by looking at the battery voltage (like you have in your second picture). It should read somewhere between 2.7 an 3.0V on battery power, and 3.2 or 3.3V on USB power.

Cheers, Harald

                  
Re: WP34S conversion, yet another but with a difference
Message #8 Posted by Gerson W. Barbosa on 22 June 2012, 7:17 a.m.,
in response to message #7 by Harald

Thanks, Harald! Your instruction was clear: "You will also have to remove LB1 and LB2". I wonder how I forgot about LB2! Well, I'll blame it on the noisy crowd that was passing by my window. No, they didn't want to see the WP 34S, they were skaters going to a skating party :-)

Cheers,

Gerson.

                        
Re: WP34S conversion, yet another but with a difference
Message #9 Posted by Gerson W. Barbosa on 22 June 2012, 9:30 a.m.,
in response to message #8 by Gerson W. Barbosa

I've removed LB2, but still 3.0 V when running on USB alone. Previously it would reset after more demanding calculations when powered on USB-1 only, but it still does when powered on a LG cell phone USB charger rated 5.1V/0.7A.

                              
Re: WP34S conversion, yet another but with a difference
Message #10 Posted by Harald on 22 June 2012, 10:29 a.m.,
in response to message #9 by Gerson W. Barbosa

Sorry, that was my mistake now. I confused it with my original design which would have shown 3.2V. What you see is OK.

About the resetting - I will look into that. I haven't seen that on my test unit. When does this happen? When you leave it running a program, or when you run a program and pres keys at the same time?

The 3.3V regulator on the FTDI chip seems to have a fairly large tolerance, maybe the voltage (after a shottky diode) is dropping below the calculators shut down voltage.

                                    
Re: WP34S conversion, yet another but with a difference
Message #11 Posted by Gerson W. Barbosa on 22 June 2012, 11:31 a.m.,
in response to message #10 by Harald

This happens when some functions are executed, trigs and factorial of large integers (greater than 170 or so), for instance. The calculator turns itself off. When the calculator is turned on again the message "Restored" appears. When the batteries are in place, the calculator turns off after a while when a program is run, but memory is preserved. This now occurs only with the LG charger (model STA-U12BS), however. No problem when connected to the notebook USB ports. The charger, perhaps?

                                          
Re: WP34S conversion, yet another but with a difference
Message #12 Posted by Harald on 22 June 2012, 6:00 p.m.,
in response to message #11 by Gerson W. Barbosa

The calculator draws a current of up to 20mA when running programs. So you should not be overloading your supply.

The calculator displaying "Restored" indicates a significant drop in Voltage. If it was just slightly too low, it would retain memory. If it doesn't loose memory with the batteries installed, that means it must switch to battery in this case, again indicating a voltage drop. This could either be due to a large voltage drop accross one of the diodes (I have measured a few of them, and they don't all have the characteristic they are supposed to have, I am wondering if I got the wrong part here..). Or it could be that your USB supply voltage drops. This seems more likely in this case, as you only seem to have the problem on the supply, not the computer. You could try adding a large capacitor to the 5V USB to solve the problem. Have you got something like 47F at 6.3V or more? If so you could solder that in parallel with C3 on my board. Positive is the pad closest to the USB connector. If that helps, your supply is the problem. If not, we need to further investigate.

Cheers, Harald

                                                
Re: WP34S conversion, yet another but with a difference
Message #13 Posted by Gerson W. Barbosa on 22 June 2012, 9:09 p.m.,
in response to message #12 by Harald

Quote:
Or it could be that your USB supply voltage drops. This seems more likely in this case, as you only seem to have the problem on the supply, not the computer.
That's very likely! I've repeated the test with another identical USB power supply, with the same results. I had thought of adding a capacitor to the USB adapter as you have suggested, but I don't have the proper screwdriver. All computer USB ports do power up the calculator perfectly, even when running an infinite loop of Zetha(3) evaluations in FAST mode, has Paul Dale has suggested. I think this closes the matter.
Thank you for your assistance!

Gerson.
                                                      
Re: WP34S conversion, yet another but with a difference
Message #14 Posted by Gerry Schultz on 22 June 2012, 9:46 p.m.,
in response to message #13 by Gerson W. Barbosa

It would be interesting to try Apple's 10W USB charger as it supplies 2 Amps of current rather than the 500mW standard to charge iOS devices, like the new iPad. Perhaps it's increased capacity would help determine if your WP34s is under powered by your USB power supply.

Gerry

                                                            
Re: WP34S conversion, yet another but with a difference
Message #15 Posted by Gerson W. Barbosa on 22 June 2012, 10:28 p.m.,
in response to message #14 by Gerry Schultz

Gerry,

The problem is with my USB charger (rated 700 mA). The problem doesn't occur when the calculator is connected to any of USB-1 USB ports (rated 500 mA) in my notebook.

Gerson.

                                                                  
Re: WP34S conversion, yet another but with a difference
Message #16 Posted by Harald on 23 June 2012, 7:26 a.m.,
in response to message #15 by Gerson W. Barbosa

As the charger is designed to charge a phone, maybe it tries to be clever and goes into some sort of power save mode when the phone is almost fully charged. The 20mA the calculator draws is not a lot compared to a charging phone. And if the charger went into some sort of burst mode, the voltage might drop before it powers up again.

Let me know if this problem occurs on a computer usb port as well.

                                          
Re: WP34S conversion, yet another but with a difference
Message #17 Posted by Paul Dale on 22 June 2012, 7:32 p.m.,
in response to message #11 by Gerson W. Barbosa

Probably the most compute intensive function on the device is zeta or the Bernoulli numbers. These would make good candidates for testing this. Even better put these in a program loop and run that. Do it all in FAST mode too to increase current draw.

- Pauli

      
Re: WP34S conversion, yet another but with a difference
Message #18 Posted by Mark Storkamp on 22 June 2012, 9:38 a.m.,
in response to message #1 by Geoff Quickfall

I don't want to nit pick, but I recall someone being confused by this in a previous thread. Based on your values for the trimming capacitor I assume you are from a country that uses . for the decimal separator and , for the thousands separator. So what you wrote for the crystal is thirty two thousand seven hundred sixty eight kilo hertz. Or nearly 33MHz. What's confusing to some is if the , was the decimal separator then you would have written thirty two thousand seven hundred sixty eight hertz, which I believe is the value you intended to write.

Quote:
timimg hardware:

-crystal..............32,768kHz, 6PF cylinder, -capacitor............ceramic 6pF 50V radial -trimming capacitor...2.8 - 20 pFd variable cap


            
Re: WP34S conversion, yet another but with a difference
Message #19 Posted by Geoff Quickfall on 22 June 2012, 11:39 a.m.,
in response to message #18 by Mark Storkamp

Absolutely correct, and not nit picking.

Corrected to north American standard radix.

32.768kHz or 32,768 Hz!

Cheers, Geoff

            
Re: WP34S conversion, yet another but with a difference
Message #20 Posted by Neil Hamilton (Ottawa) on 22 June 2012, 12:04 p.m.,
in response to message #18 by Mark Storkamp

Back in the day when I used to do board design and specified components, we would use the power of ten designator as the radix mark.

For example, this crystal would have been described as 32k768 Hz (or 32K768 Hz if you prefer).

It seemed to work for us though I am not sure how universal it was.

                  
Re: WP34S conversion, yet another but with a difference
Message #21 Posted by Harald on 22 June 2012, 6:02 p.m.,
in response to message #20 by Neil Hamilton (Ottawa)

Exactly how I do it. Very common in Germany and the UK.

      
Re: WP34S conversion, yet another but with a difference
Message #22 Posted by Katie Wasserman on 22 June 2012, 10:34 a.m.,
in response to message #1 by Geoff Quickfall

Geoff,

Great pics, neat work and nice writeup as always!

What's your maximum IR printing range? I've converted a few machines and found some variation even using identical parts and methods. The best is around 18 inches, the worst about 10 inches.

Good luck with your crystal trimming adjustments.

-Katie

            
Re: WP34S conversion, yet another but with a difference
Message #23 Posted by Geoff Quickfall on 22 June 2012, 12:28 p.m.,
in response to message #22 by Katie Wasserman

Thanks Katie!

Upper range is 24cm, according to the 34s conversions that is 9.5 inches. I will get some new batteries and try again, but the voltage is 3.0.

I tried shims to change the angle of the calc, correcting for LED positioning and the best I got was 25cm. Indicates that the IR Is either positioned correctly or doesn't care.

I am going to do a second 34s but I will route the LED to the side. This will allow me to position the calc in the 'office' sideways to the printer allowing me to see the calc correctly. This is due to the table constraints in the cockpit. I have always wanted the 41/42 to be sideways to the printer at work.

Cheers, Geoff

Edited: 22 June 2012, 12:31 p.m.

      
Re: WP34S conversion, yet another but with a difference
Message #24 Posted by Marcus von Cube, Germany on 22 June 2012, 2:54 p.m.,
in response to message #1 by Geoff Quickfall

Quote:
Special thanks to Pauli and Marcus (his mother makes good cookies)
It was her mother, mine isn't any more.

Thanks for the excellent description.


[ Return to Index | Top of Index ]

Go back to the main exhibit hall