• Please review our updated Terms and Rules here

reproduction PDP 8e toggle switches?

firedome

Experienced Member
Joined
Aug 20, 2013
Messages
197
Location
NY
Is/has anyone reproduced the PDP8/e orange (two color tones) plastic toggle switches?
Don't have cad or 3-D printer access, so hoping someone already has done this and has these available,
they seem to be a commonly failed part...
thnx Roger in NY
 
BTW: I've one of Rod Smallwood's reproduction front panels and it is superb. I'm no expert but I cannot tell any differences from the original. Not sure if they are still available, but if so get one while you can, highly recommended!
 
If its just the plastic nibs holding them in place are broken you can drill and insert a metal rod to fix them. The colors don't seem to match well between different machines so replacements may not match well.
 
Thanks, no, I have an 8/e light/switch PCB panel with multiple missing plastic switch paddles, so not a matter of repair.
Imperfect colors are ok, better than no paddle at all.
 
I have asked for the same knobs a while ago...
And I have tried to reach mr Rod Smallwood. But he does not answer any emails...

So I have been talking to a specialist which I know personally. He is a well known professional in the replica business for years.
He has reproductions of knobs, dials, and many other parts. https://www.bendijkman.nl/

I have a price for PDP8 front panels already and he is investigating the knobs. Because materials have
always a bit colour difference, the idea is to make a complete replica knob set for one machine.

The point is, that it is unknown how big the demand will be. And then what about the costs?
How much would anyone pay for a complete set of replica knobs for a PDP8?
If that is known Ben has an idea about how much effort to put in these replicas.

So maybe we should have a poll on these knobs. Who is interested and how much are people willing to pay for it...

Regards, Roland
 
Last edited:
From the look of it, these toggle switches are not very complicated and should be easy to design in any CAD tool. And using SLA 3d printing, you can produce lots of them in little time and in perfect quality. Color is best done using spray paint later.

Too bad I don't have a PDP 8e toggle switch as a blue print, otherwise I could help out.
 
My son is a 'whiz' with 3D printers - he used to work with them for a job.

Things are not as simple as you may think... These took a long while to produce - they have a high-quality surface finish.

Anyhow, he designed some PDP-8 switch toggles for me for a project (see https://drive.google.com/drive/folders/0ByKv0MbpuX4CeEUxZElDYjdoeG8?usp=sharing). Not quite identical to the original - but suitable for my use. He developed two designs - one with the spigot and one without. The problem with the spigot variant was one of resolution and strength. The ones with the holes for a rod are beautiful.

The orange and yellow colours are filament that is 'off the shelf' - so a little garish... However, you can get colour-matched filament now. Not sure what price we are talking about though.

Willing to share the design if that would be of help.

Dave
 
Things are not as simple as you may think... These took a long while to produce - they have a high-quality surface finish.
Not sure why the "high-quality surface finish" of the original switches matter when talking about re-creating them for 3d printing. In case you refer to some kind of texture, it takes 5 minutes to make a grid of very small spheres and substract that from the outer surface of the 3d model to create a texture. I think many people overestimate the complexity of modern CAD tools. All that stuff has become very easy to do.
 
Yes, it is a derivative of that part.

The 'high quality surface finish' I was referring to was for the 3D printed item, not the original.

I have seen quite a few 3D printed parts where the quality of the final result was awful...

Dave
 
Is/has anyone reproduced the PDP8/e orange (two color tones) plastic toggle switches?
Don't have cad or 3-D printer access, so hoping someone already has done this and has these available,
they seem to be a commonly failed part...
thnx Roger in NY

This is a perrenial question that comes up somewhere every year or so. Probably because no readily available high quality solution exists (yet?).

The first challenge is to get something that works like the original -- that lines up properly, etc. There are a number of folks with perfectly fine CAD files that will 3D print something functional, including mine:
http://www.so-much-stuff.com/pdp8/cad/3d.php
If you need small volume and don't want to 3D print a finicky part, you can just order them from Shapeways, etc. That addresses "fit".

"Finish" is a harder problem. The 3D parts have surface texture, even if it's just the matte finish from sintered nylon. (I haven't seen a resin print that is strong enough.) One approach to solve this is to make molds for cast resin parts. This is tractable, and I've been able to make molds that will cast good parts most of the time, even without a vaccum to help remove the trapped bubbles. I reccomend casting about a baker's dozen at a time, both to deal with rejects and to help with color matching.

Color matching is a bear. It is nearly impossible to reliably get a consistent color from batch to batch because of the small volumes of dye involved and such. Not to mention that what constitutes a decent match seems to vary with lighting conditions, etc. etc. The only thing I know that will help with that is to make a whole color set at once, from the same batch of the same material. (I have never pushed it hard enough to get to where I could reliably do this.)

I think it might have been Dave G. who years ago had a handful of batches made by a proper injection molding company, and I have one of those. The colors aren't "right" to my eye, but if you replace a whole panel no one is likely to notice. That's the closest "fit and finish" that I've seen.

Another quibble: I agree that the hinge pins can be replaced to refurbish a handle by using a drill press to drill through the handle body, and installing a rod. I argue that metal (often a straightened paper clip) is the wrong material for the new axle. I want the new axle to buckle if the switch handles are impacted, rather than damage the even harder to replace slide switch underneath. I've had good luck making replacement axles from wood, etc. The result is inexpensive and easily replaced. I can get 1/16" square sticks 3' long at my hobby shop. It only takes seconds to sand the piece you need (just over half an inch) round and install it. (Sand before cutting, so you've got a place to hold onto it. Spin the stick in the abrasive until it is round.)

Vince
 
I have seen quite a few 3D printed parts where the quality of the final result was awful...
Mostly because of the use of FDM. Printing these with SLA or DLP will probably result in a better quality than the original parts.

If the thingiverse.com model is the correct one, I'll do a test print with my DLP printer and see how it goes.
 
Last edited:
There are several issues going on here at one and the same time.

1. Original switch durability. The other day I was trying to hang a painting on an interior wall. Piece of cake. Drill a hole. Install a plastic anchor. Install screw. Hang painting. But the only anchors on hand of the correct size were in a yellowed plastic package that was already cracking. The anchors themselves tended to disintegrate when i tapped them into a pre-drilled hole, or when the screw was installed. The plastic was just old. Chances are good that if you need to replace one or two of your 40 year old switch handles, they're all on the edge of coming apart. Color matching new handles in onesies and twosies as time goes along might not yield the results you want.

2. Original switch color. Even if you manage to exactly match the color of your particular switch handle on your pdp8, chances are good that your original handles are no longer the color that they were when they left the plant. I would also guess that there were color variations on new production handles from year to year and plant to plant. So a match for your machine might not be a match for someone else's Given those two issues, if i were restoring/repairing a pdp8, I'd do the whole set in one go if at all possible.

As for the issues raised above on availability, quality, and cost of repop parts, someone is already dealing with those issues for IMSAI restorations. What they're doing and how they're doing it might be a guide going forward on DEC switch handles for pdp8's and pdp11's.

https://www.ebay.com/itm/4-Replacem...192265?hash=item34268ecd49:g:fQQAAOSw59hen25Z
 
As far as durability, I suspect that something like PA6 glass-filled nylon would have no problem lasting several decades. But that's beyond the facilities of a typical hobbyist.
 
1. Original switch durability. The other day I was trying to hang a painting on an interior wall. Piece of cake. Drill a hole. Install a plastic anchor. Install screw. Hang painting. But the only anchors on hand of the correct size were in a yellowed plastic package that was already cracking. The anchors themselves tended to disintegrate when i tapped them into a pre-drilled hole, or when the screw was installed. The plastic was just old. Chances are good that if you need to replace one or two of your 40 year old switch handles, they're all on the edge of coming apart. Color matching new handles in onesies and twosies as time goes along might not yield the results you want

In my experience the switch failures are not due to age. They come about because someone thinks they can carry the 8/e chassis by themselves. You either pick it up with the switch panel pressing against your stomach or you try to pick it up from the rear and realize it would balance better if you picked it up from the front and then switch. The problem is that the molded pins are not super strong and they shear off in a moderate impact. If it is carried from the rear then the tendency is to bump something against the switches and break the pins. Once you repair it, always get someone to assist in doing so. They are just a little too large, heavy, and fragile to risk carrying by yourself. I would suggest that an overall time saver would be to remove the cards and front panel if you are going to haul it around in a vehicle. You have a lot better chance of it all working when you get there than if it gets dropped or banged up.

You can FDM 3D print acceptable handles but not if you try to print the pins. Modeling the original part that was intended for injection molding also doesn't work all that well. The model on Thingiverse is pretty faithful to the original but prints poorly at best. I did a model in sketchup a couple of years ago that I printed with the top face of the handle on the glass bed and they look better than the originals. I should re-do the model in Fusion 360 and work towards making it even more printable than the earlier one. These were for my DECSet 8000 which as far as I can tell is the most rare of the 8/e variants. The front panel colors are midnight blue and sky blue. I found good color matches in PLA. I don't have an 8/e to try to color match and it is a futile thing to chase. If there was actually a market for even a few thousand switches it would be worth ordering a custom color match. At that time you needed to order a minimum of 20 kg (44 lbs) of filament in each color. That is a LOT of switch paddles. I will attach the .stl file if I can find it. I don't have the sketch up original.

Doug

The forum does not allow .stl so I renamed it .txt. STL files are text so this is still technically correct. Lets hope it comes through anyway.
 

Attachments

  • DS8khandle.txt
    19.2 KB · Views: 3
Last edited:
In my experience the switch failures are not due to age. They come about because someone thinks they can carry the 8/e chassis by themselves.

I've also had mine "safely" in the vehicle, only to have the load shift as I went around a corner, and pressing up against whatever was enough to snap the pins on a couple of handles. Sometimes I am tempted to purpose-build some sort of shield for them, but I haven't done that yet.

One time I did ship one, and then I did use rigid foam IIRC to form a "box" over the switch handles.

Vince
 
The 8e toggle switch paddles look the same as on my 11/15, shown below. There is a prominent mold line artifact down the middle of the paddle, and the two halves are a tiny bit disjoint.
This is of course exacerbated by the amount of dust and grime on them :). This "feature" could be added to the CAD model perhaps.
FOX_2_switch_paddles.jpg
 
It seems plausible that the eBay seller of the IMSAI paddles would be in a good position to duplicate the PDP/8e/f/m, PDP/11/x switch paddles given he's evidently had success with those he's selling. In order to explore this possibility, consequently I've sent
Don Bailey a message as below, and will report on his response. Worth a try in any event:

"Hi Don:
There seems to be a quite large demand for very similar replacement paddles for DEC PDP/8e and PDP/11 machines, it's been a topic of interest in the DEC community for some years. Would this be something feasible for you to do, or perhaps be able to let us DEC folks know how, or by whom, your paddle reproductions have been accomplished? I believe that pre-production orders for them would be significant. Cad files have been created for the shape but materials and results used thus far with 3-D fabrication have been less than satisfactory. Please advise and thanks.
thanks,
Roger Anderson
Endicott NY"

Another possibility occurs to me. Not knowing much about plastic molding or metal fabrication, I wonder if these could be duplicated in machined aluminum. A good friend who retired from IBM fabrication has a beautiful automated CAD/CAM machining system in his garage. Some years ago I commissioned him to fabricate NLA machined aluminum knobs for early 1960s Harman-Kardon Citation audio equipment, which have been successfully produced and sold for over 10 years now. These require both machining and broaching. Once the designs were completed using CAD and programmed in the cost came down as numbers increased, to an average of about 10-20$ depending on size. Now, I have no idea if the paddle design is amenable to this type of process, but if so it should certainly be durable. DEC should have made them that way from the very beginning for so expensive a machine imo! So, anyone who knows more about the subject, please chime in as to if this might possibly be a feasible approach. A finished machined aluminum paddle would then require only paint matching for correct original color.

From what I can see in the responses above, a good number of folks require paddles in numbers significant enough to explore their reproduction further.
Roger in NY
 
Last edited:
Back
Top