Musings and comments about our common interest
We are in the final stages of writing a book on HP41 repair. We expect to have it finished by the end of September. In our experience you do not need to have any electronics knowledge to repair 80% of HP41's damages. The nature of many of these is either mechanic or bad connections due to corrosion. The focus of this book is on that 80%. It is written for those of you that wouldn't dare to repair your calculator before!
Here are some of the chapters:
The target price will be 9,99€ for the kindle version a 2-3 € more for the print version, and will be sold through amazon.com and on this website. I'll keep you posted through this channel. Do not hesitate to bring suggestions for additional content!
We have introduced a new product. This is only for those that want to use the HPCL circuit on an old HP41c (with serial number before 1954AXXXXX). The screen set requires that the processor have specific capacitance. This is provided by the processor that came with the machines, but not by later processors (in case the original became defective) or if you want to install a CL super-processor. So we have bought a pile of 10nF surface-soldered capacitors, to install between two parallel circuit lines. We will provide instructions if needed:
We have done more tests and now we have put it for sale: it works well. Now thinking about packaging to send them in the best way - so that the adhesive stays put until installation. Alternatives are mounting it on a lower post assembly, or another clipping method to maintain the pressure between up and down.
Further pictures on the first production run:
There are cases (black spot on the screen, screen only works for half the digits, broken screen) where the only solution is to change the screen. These are not produced anymore, so you will need to find a donor HP41c.
To replace the screen: First unsolder the two screens (from the donor and the damaged calculator) Use a small soldering gun and help lifting the pins with the tip of a plier, one by one. Important: unsolder the screen at the calculator side: it is much more resistant!!!
Then solder the donor screen back in place. Locate the screen in its hole first. then start soldering pins, with a very clean soldering gun. There is no need to add more solder. After melting the solder, leave the gun while maintaining pressure on the pin with a flat screwdriver tip, to ensure good, durable contact. Patience until finished. Then reassemble and try again.
I have received today a fantastic tool for testing the HP41c/cv/cx calculators that I usually repair: clones of the HP service modules.
HP released two versions of the service modules: one for HP41c and another for HP41cv/cx machines. You can read the details from the service manual:
These modules greatly simplify testing, including memory, screen, keyboard, status, etc. We will discuss these modules and other diagnostic tools in further blog issues.
If you remember, we had ordered a new time of 3D printed battery modules, specially designed to be used with glue on broken calculators back sides:
This version is intended in case of both supports of screw head broken. For any other case, the standard piece would fit better. The repair is also definitive: if it doesn't work or gets corroded, the repair would be much harder!
Here is the body for the standard piece:
The pieces have arrived and we will do some tests during this week end. We'll come back with the results next week!
We've just received three calculators for repair: a CL that worked on assembly but not on delivery, a C and a CV. The C is easy to diagnose: corroded battery terminals. Let's see if it is the only thing needed when we open it. The CV has a beautiful battery terminal and looks solid, without any post broken - but it doesn't work!
It will be an interesting week end...
I'll try to take some pictures to illustrate further bolg issues - also with oscilloscope output views, to compare a new CL with older models, when it comes to wave frequency stability.
Please contact us at firstname.lastname@example.org if you need your HP41c serviced!
We have returned from holidays with a new 3D design, although unrelated to calculators - a battery door for the ubiquitous Samsung TV remote!
We will have it in different colours - as a way to distinguish different remotes. It is still not created as a product, but it is on its way.
We have several products created as 3D: here is the list so far (not counting repair parts):
- Samsung TV remote battery door (coming)
We are preparing the development plans for 2020. Your requests would be listened to, but be aware that someone insisted heavily on creating the Woodstock battery bay and we didn't get an order from her finally! So we would like to have some quorum for the demands.
My list so far:
- USB connected HP41c adaptor (not battery - putting any kind of electronics inside would require huge certification effort (or alternatively huge risks for our very small company). It will be a 3D printed part with a connector. The jury is out as to whether to include a capacitor to hold the charge for longer.
- A permanently glued battery-ports connector, to hold better together with the back case.
- Many other small items for HP calculators: screws, spacers, screen plastic covers, etc.
- Finish a HP41c repair manual (we're already half way through)
Please also submit your requests to email@example.com too!
When repairing a HP41 machine, it is always interesting to know what it is happening inside - regardless of the screen being blank or not. Using an old memory module, we have connected the pins with a ribbon connector, and then this connected to the oscilloscope we can see several things: is there the positive voltage? when you press a key, does it appear on the interface?
The system is designed so that when a key is pressed, there are two lines ( ø1 and ø2) that output a synchronizing wave - delayed some time the one versus the second. This is a tell-tale that the circuit is working - regardless of what happens to the screen.
Here you can see in the case of a CL just installed (thanks Mr. Lerich). The circuit is outputting 360 Khz, much more stable than the usual 343-365 that you see on normal calculators. The ringing at the corners may be due to the fact that the cables were not terminated properly and therefore there were reflections. Also it can be seen the induction effect on the rises of one curve on the other, due to the parallel 150mm of the ribbon cable.
If we see it closer, we see the rise-time of the curve is equivalent to 25 MHZ - quite a high bandwidth, even despite the mentioned termination issues. Also the stability of the voltages seem very good. Both are testimony to Monte Dalrymple's good engineering!!
We need to get deeper in this method to be able to better debug and repair our calculators. Now we need to get a service module - or an image burned into an eprom of such.