Musings and comments about our common interest
We are now designing an adaptor to put two AAA batteries in a Spice machine. The Spices were the HP32E, HP33E, HP33c, HP37E and the HP34C - the top of the series.
The Spice machines were designed as low cost compared with the HP65-HP67 series, and as replacement of the Woodstock series. In opinion of many HP experts, too many corners were cut to manufature them - several internal connections rely on pressure exerted by the screws that close the box - while in previous units the box screws did not have circuit integrity functions. Therefore, these machines have higher fatality rate than other models.
One of the issues that plague the HP Spice is the battery bay. The original rechargeable batteries wear down, but the problem for the replacement is that the normal AA size is 1-2 mm too long for the machine - and you break the machine in the medium term. Ignacio Sánchez suggested to design an adaptor that used two AAA batteries - where the size challenge does not exist.
(continuing with the story: after the Spice series, the lower range was inherited by the Voyager series - for many, the most perfect calculator design ever, with the HP12c and the HP15c as the main examples)
We have applied the same solution that we created for the lower posts to the upper posts: a grip that clips on the keyboard circuit, and strengthtens the grip with the introduction of the screws.
In the case of the upper posts there was a requirement to drill anyway - so it doesn't change the original procedure. But the piece has the same 3-petal grip that we use for the lower post:
The connection strengthtens when the screw is put in, since it makes pressure to the sides of the post and makes sure it doesn't move. We plan to put it in production already!!
During the last week I have tried to repair (or install CL circuits on calculators) many calculators: 17 of them, of all types and flavours!
Again, repair statistics are quite stable, but there is a jump between 1st and 2nd:
All in all, a good week where we have managed to repair 13 of the units, there are still 3 to go and one completely unserviceable! Some have already been sent to their lucky owners.
Now I have a pile of calculators to fix. Many of them have corrosion points in the circuits and -above all- in both sides of zebra connectors.
No matter how thick the corrosion layer is, our recommendation is to put some drops of Caig DeOxit 100% (although other formulations (5%) may work too) and let it act for a little while. Then you can remove the crust of corrosion with a flat small screwdriver. You will be surprised because most of the crust will go and you will see the yellow colour of the copper below!
There are some areas where the corrosion is more difficult to remove: the parts where the main current happens. The additional resistance that builds up with the initial corrosition creates some heat that burns the debris in the metal path and makes it much more difficult to remove. One of the main causes is the "dust" that comes from the disintegration of the fabric inside the vinyl cases.
(By the way: it is a good practice to wash the vinyl cases to remove the dust. Use normal soap and doit by hand)
This product barely leaves any residue, and although it is quite expensive lasts a lot (if you use it drop by drop).
This product is typically used in audiophile circles to improve the conductivity of the Hi-Fi connections. I discovered there but it can be used in any case the connection is not good. It is one of the products you feel proud to recommend, like in "you will thank me to bring this to your attention.
We have detailed with pictures in our book how to repair the most typical card reader repair: the "gummy wheel" problem. This appears when the original material that was used to move the card disintegrates into a gooey mess. This is covered both in the book but also on a number of other websites.
We have found also another type of damage: a similar disintegration in the material between the engine axis and the endless screw that transfers rotation to the wheel. It shows in the noise of the transprt (which is the endless screw skipping over the engine axis). This will appear in the next version of our book. It basically consists on:
I will include pictures once I have done another fix.
Some interesting units arriving! Some will need repair, but altogether very interesting details! a family picture:
We have been using for already a couple of years our lower post repair piece. It is designed to replace both lower posts of the HP41c, since these are often broken in the full-nut HP41c clculator, and which are key for the calculator work since they hold in place and under pressure the nut processor circuit. The idea for the part was to have a single piece for both posts - the attachment on one side provides torque resistence for the other side and vice-versa.
This has been quite a success but it requires good alignment of the piece with the original holes, and it depends heavily on the glue quality. If the pulling force was too strong, we risked to have the piece off the keyboard circuit. This is despite having given a wide gluing area.
We have created a new low post repair part that requires some drilling of the rest of the original post, and introducing "claws" inside the hole. The piece is designed so that when the screw enters in, it pushes the 3 "petals" of the claw to clip below the keyboard circuit - making it much less dependent on the glue!. It is really strong - the longer the screw the stronger it is. Definitely an advancement over the previous method. What you lose having to drill deeper the original posts, you gain ensuring a perfect alignment, and a much better resistance to pulling forces.
Here are some pictures of the piece (sorry for the bad lightning conditions!) and 3D-renderings too:
We have been asked to produce new HP IL cables, and we weould like to know if you are interested.
From a technical point of view is not too difficult - the extreme low speed of the nework make it very simple to make it work - there are no complicated microwave effects to solve. (Hey, were talking of less than 50kHz frequency!!). The issue is to get the connectors, both housings and metal. In particular the metal parts are quite expensive: close to two euros each metal contact, of which each cable pair has 4! And then there is some soldering and finishing required.
The question, of course if you as customers may be interested in it. Please respond to this thread, or send a mail to firstname.lastname@example.org with your opinion.
By the way: we do have an HP-IL printer if needed. Please remember that if you have the HP-IL module, it doesn't consume an additional port, but instead can be dasiy-chained tothe other items in the HP-IL loop!
We have designed some changes to the battery assembly. The main of them is to create a small dip in the main body so that the band that maintains the film in place but that doesn't interfere with any kind of battery holders (we have seen some cases where there was minor interference.
Here are some pictures: