• Please review our updated Terms and Rules here

Mini-Omnibus backplane for debug and minimal system

I took Vince's switch paddle and played with it until I could get it the best that I can on my Prusa MK3S+.

Here are the results:

https://drive.google.com/drive/folders/1ogop3aTsTOpLXn-rV4LReY2V4WLvuCEy?usp=share_link

This were printed standing on the tabs that slide over the switch on the circuit board and printed with supports with Prusa PETG orange filament.

I don't have the DEC colors but if anyone wants to send me the filaments in the DEC colors I will print them a complete set.
 
BitWiz said:
I took Vince's switch paddle and played with it until I could get it the best that I can on my Prusa MK3S+.

Here are the results:

https://drive.google.com/drive/folders/1ogop3aTsTOpLXn-rV4LReY2V4WLvuCEy?usp=share_link

This were printed standing on the tabs that slide over the switch on the circuit board and printed with supports with Prusa PETG orange filament.

I don't have the DEC colors but if anyone wants to send me the filaments in the DEC colors I will print them a complete set.
Click to expand...
After having some parts including 8/e switch toggles professionally printed by JLCPCB I haven't used my Creality Ender 5 as I cannot even remotely match JLCPCB's quality or the choice of printing materials.
Likely there are local companies who can produce the same or similar quality without the shipping delay & cost.
 
Once i got the orientation right these came out pretty good but to be honest I think an SLA printer would do a better job than an FDM printer.
 
pbirkel@gmail.com said:
Can you tell me the shaft distance from the shoulder where the knurling ends to the base of the knob?
Click to expand...
0.387 to internal base of knob. 0.280 to skirt of knob. My knob is stiff pushing in so I don't push it in very far. You can see in the picture its sticking out quite a bit. Suspect originally it pushed in further. You may want to adjust diameter or length to fit better.
 
djg said:
0.387 to internal base of knob. 0.280 to skirt of knob. My knob is stiff pushing in so I don't push it in very far. You can see in the picture its sticking out quite a bit. Suspect originally it pushed in further. You may want to adjust diameter or length to fit better.
Click to expand...
Thank you for the additional dimensions. I'm not sure what material to use for a replacement shaft; I confess to thinking seriously about a hardwood dowel as I think that might "take" the knurling without any attempt to tool a formal knurl on the end. Next choice would be a plastic rod on the same basis but I suspect that anything that would be soft enough to form may subsequently de-form and slip. Once I have the end off the switch I'll be better able to work on fitment. And I'll be better able to assess the castellated end; perhaps a "full length" permanent shaft will be appropriate to provide more strength ... although it's not clear to me why that wasn't the original design for the switch. Must be a good reason that I haven't thought of yet.
 
BitWiz said:
I took Vince's switch paddle and played with it until I could get it the best that I can on my Prusa MK3S+.

Here are the results:

https://drive.google.com/drive/folders/1ogop3aTsTOpLXn-rV4LReY2V4WLvuCEy?usp=share_link

This were printed standing on the tabs that slide over the switch on the circuit board and printed with supports with Prusa PETG orange filament.

I don't have the DEC colors but if anyone wants to send me the filaments in the DEC colors I will print them a complete set.
Click to expand...
Thank you for the test sample & photos. That's a nice orange! Of course they show the usual striations which might be reducible using the acetone-vapor technique -- or so I read; I have absolutely no hands-experience with the technology.

For me your sample falls into the "uncanny valley", so to speak -- close to original but with striking "I'm a poor substitute" visual characteristics that I would find distracting. I'm leaning towards the well-engineered replacement from JLCPCB that doesn't pretend to be the correct color but otherwise has great fit, finish (mostly), and durability.

I tried pricing a PDP-11/70 switch handle using the same process and it came out to >$4 each. Lots of volume to fill in that triangular prism I guess. Perhaps that design needs to be reworked to something more honeycombed (<20% fill?).
 
gwiley said:
The front panel does wobble a bit in the mini backplane. However, I've discovered that I don't go to the switches much on the front panel since building the M847-Extended boot loader.

True, and I've been thinking about making one for the left side as well. However, that's been delayed because I've started designing a jig to support the front panel plus plexi-panel. A left-side card guide would be a quick project though.
Click to expand...
I'm interested in seeing where your jig-design takes you :->. Is your plexi-panel original or a reproduction from "panelman" (Ron Smallwood)? Are you installing/using a light shield around the lamps/LEDs?
 
pbirkel@gmail.com said:
I'm interested in seeing where your jig-design takes you :->. Is your plexi-panel original or a reproduction from "panelman" (Ron Smallwood)? Are you installing/using a light shield around the lamps/LEDs?
Click to expand...
The plexi-panel is a reproduction and the "spare" front panel board is the reed switch type, 180 degree rotary switch (same as yours). I've removed some switch handles from this spare board to fully-populate my main system. (So that system in a real chassis looks nice.)
The missing switch handles on the spare front panel board have been replaced with 3D printed switch handles that you saw in the photo in a prior post. This front panel will go into the mini-backplane system for debugging, so functionality is the most important thing. I was thinking about alternating black and gray switch handles, or maybe just replace all of the gray switch handles with black ones.
I'm using this for the knob on the rotary switch: https://www.amazon.com/gp/product/B07F25QGDN
It feels a bit too modern for a 1970's computer but it's functional.
 
thunter0512 said:
Why did you chose to have to card guide only on the right side rather than on both sides? It should be relatively easy to just mirror the design.
Click to expand...
I made a short and tall version of the left side card guide. The tall one seemed like a good idea but the height to mounting foot length ratio is rather large so it bends out of the way easily and the cards can slip out of the guides more easily than I imagined. I think the short version might be better.

There will probably be some more 3D files so I created a new repository for 3D parts: https://github.com/G-Wiley/Mini_Omnibus_Backplane_3D_Printed_Parts
 
Thanks George. I made happy-snaps of two of your STL to give folks (especially me!) a general idea of the full-10 and then front-notched 9-only brackets.
 

Attachments

  • Mini Backplane Card Guide 10 same left short v02.jpg
    Mini Backplane Card Guide 10 same left short v02.jpg
    29.5 KB · Views: 6
  • Mini Backplane Card Guide left short v02.jpg
    Mini Backplane Card Guide left short v02.jpg
    29.3 KB · Views: 6
thunter0512 said:
After having some parts including 8/e switch toggles professionally printed by JLCPCB I haven't used my Creality Ender 5 as I cannot even remotely match JLCPCB's quality or the choice of printing materials.
Click to expand...
I finally ordered a set of switch handles as well as a pair of card guides from JLCPCB, both SLA. They arrived last week. WOW, just WOW. At least for me the finish of SLA is truly amazing. Every bit as nice as that photo you put up, if not more so. Will have to see how the pivots hold up over time but they look and feel durable, perhaps even more so than the originals.

The card guides are quite solid, literally. Probably could benefit from some honeycombing to cut mass without affecting rigidity between the PCB slots. But as a one-off design I think that they are great. Thank you George!

BTW, I also printed a test-case 1170handle from so-much-stuff (VRS). Got grumbling back from the JLCPCB QC process about strain areas in a few places (see attached illustration) but they eventually agreed to print it (SLA 8111X Resin, white) for me. It's marvelous, with same caveat about seeing how well the pivots hold up. Will include a fairly complete set in my next order. I'm sold on JLCPCB.

Finally I ordered a pair of so-much-stuff (VRS) front panel brackets as another test case; I used SLA 8228 Resin, light green. These are more flexible than the other prints, which is desirable in this application. I tested against a few detached front panels and my impression is that the pins are placed slightly further apart than "exact". I need to try mounting them in a rack where the main strut won't be able to flex as much and see how well they work in practice.
 

Attachments

  • 1170Handle Eval Image.png
    1170Handle Eval Image.png
    303.8 KB · Views: 4
Last edited:
gwiley said:
I'm using three NicePower adjustable bench supplies for +5 +15 -15, and a 9V power adapter (found in my junk box) for to power the 8/E front panel lamps (the "yellow" wire). I'm kind of nervous about not having OVP in this present temporary configuration.
Click to expand...
What did folks decide for their final power supply solution?

For reference, the H724 specs are: +5Vdc +-3% 20A, +15Vdc +-10% 1A, +8Vdc +-25% 2A (panel lamps), and -15Vdc +-5V% 8A.
 
Is there a document listing power requirements for various modules/internal options? I've been looking up the requirements for -15Vdc as regards the various MM8 and and am not finding any specifications. The MR8 ranges from 100mA to 300mA for the -15Vdc line, but that's the only module that seems to have a specification. I'm wondering if I stay away from core memory I could use an ATX SMPS and simply boost to +-15Vdc used typically for serial interfaces and low current requirements. Then substitute a temporary -15Vdc high(er) current supply if running core? Just thinking out loud here ...
 
The H724 powers two full Omnibus backplanes.
The fully populated mini-Omnibus backplane will draw substantially less power. The Meanwell power supplies from post #4 are more than sufficient.
For the 8VDC supply I used a cheap 9VDC Chinese supply from Aliexpress. The supplier I bought from has none left, but there are plenty of others. On Aliexpress search" "Switching Power Supply 9V".
Tom
 
Agree; dividing the -15Vdc output current in half and then populating that backplane with a pair of MM8-x, each with a max-draw of 1.5-2A sounds "about right" based on somewhat similar PDP-11 core memory assemblies. I don't actually have any quad-height core memory -- it's all hex-height (MM8-AA / MM8-AB), so presumably no-go with this backplane anyway. And if I should happen-into a quad-stack (MM8-E or MM8-EJ) then 3A ought to be adequate for that and an RS-232 serial module/etc. The front panel that I expect to rescue is LED so the+8Vdc requirement is probably inapplicable.
 
I built a small debug setup that's quite similar to Tom's system. Wanted -15V to be rather robust to debug core memory. +15V is minimal.
+5V 10A: RS-50-5, Digikey 1866-4157-ND
+15V 1.7A: RS-25-15, Digikey 1866-4141-ND
-15V 5A: RS-75-15, Digikey 1866-4159-ND
The total was about $54 plus tax and shipping.

The +8V for the front panel lamps is still a spare 9VDC power adapter plugged into a separate outlet. In my 8/m chassis that has an 8/e front panel I'm using a small DC/DC converter to generate +8V from the +15V output of the H724.
Should probably do the same with the mini-backplane system.
 
Thank you George; that looks like a really nice set-up. Nice boost-buck converter with display too!

I've been scrounging through my odds-n-ends and found an open-frame 15Vdc 3A SMPS as well as a 5Vdc 20A SMPS (overkill!) and an appropriate DC/DC boost regulator that I could use for deriving the +15Vdc rail. Still on the fence over whether I'd be short-changing myself on the -15Vdc rail :-<. Decisions, decisions ...
 
You must log in or register to reply here.
Back
Top