MK14 schematic revisions

[philoupat83]

hi all finished editing my MK14 keyboard in 3d printing everything works with XOR the keyboard and not top I prefer my PIC14 for programming

[SiriusHardware]

Excellent implementation of the keypad frame and buttons!

How do they feel when you press?

PIC14 is good to develop MK14 code on because it has serial code download included.

[philoupat83]

thanks
I think that still needs to change the shape of the keys still not cool bounces
yes I like the port serie a programming in assembler compilation pouring and voila
a great thing for education thank you karen

[philoupat83]

I palm my 3d keyboard and I would like to add a connector for a future VDU which connector?

[SiriusHardware]

The original VDU is made to use a DIN card rack connector.

See post #133 of this thread for images of a VDU PCB fitted with that type of connector.

There are also further discussions in this thread, post #167 onwards regarding the best way to connect the VDU to the MK14 rear connector.

Your Martin replica probably has contacts on the lower side of the rear edge connector. You can wire the required system Address, data and control signals to these so they are available for the VDU.

Your JM replica PCB probably does not have contacts on the lower side of the rear edge connector.

Therefore, the 'Martin' PCB is the best one to use with a VDU if you have both versions.

[philoupat83]

thanks

[SiriusHardware]

I've been involved in a conversation in another forum with someone who had a newly built 'Martin' replica which wasn't really resetting properly with the reset switch, although he discovered that if he disconnected the reset capacitor, the board would reset properly with the switch. But then it did not reset properly at power on.

The fix he arrived at was to put a 0.1uF across the reset -resistor-, and now his Martin replica resets properly on power up and resets properly with the reset switch. I seem to remember Tim and Slothie both had problems with 'dirty' reset on their JM PCBs, so it would be interesting to know if this is also a 'fix' for the JM PCB.

There's an added complication, in so far as the JM PCB probably uses the original 'reset switch down to 0V, two gates' scheme whereas the Martin PCB has one less gate and the reset switch going up to positive.

Do the Martin and JM PCBs have plenty of decoupling capacitors on the 5V supply, or are they both similar to the original MK14 with almost none?

[Timbucus]

As with the original they have little or no decoupling - in that sense the JMP ones are exact replicas of the VI/II boards. As you say they have the traditional pull down circuit as well. My reset is pretty reliable. The reset issues we had were with the PI programming board and its timing of a reset rather than the board itself - but, I suppose if the CAP charges too soon it will being the power up from the zero level which was what was happening for me.

[SiriusHardware]

Yes, I certainly do remember you having to twiddle with the reset timing on the uploader and come to think of it the machine Slothie had reset problems with was actually his own original replica.

Originally Posted by Slothie

....The only odd thing is that on reset 8 out of ten times you get random digits on the display in the gaps between '0000' and '00' - but pressing Abort clears this and all seems to operate as normal.

I don't know whether he ever sorted that.

[Slothie]

I worked out by experiment that discharging the power supply electrolytic made it reset properly on power up. I thought the problem might be power supply noise as there was a lot, adding a few decoupling capacitors seems to have fixed the problem.

[philoupat83]

Hello
on my retraction martin no problem of reset my supplies with a stabilized power supply without decoupling to the old so no harmonic in the regulator and the filter capacitor or that of the reset

[SiriusHardware]

My parts collection for my 'Slothie' replica MK14 is one step closer to completion, as the RAM is not far away now. Unfortunately I have had to pay Royal Mail £6 + £8, some of that Customs duty (not unexpected) and some of it 'Handling Charges' (??) in order to get it delivered. That has added quite a bit to the otherwise modest cost of the parts.

Now only requires key switches ... and a PCB to build it on.

[Slothie]

It's a shame because that handling charge is for collecting the duty! I think my RAMs came duty prepaid which means you save the handling charge, but the postal charges were suspiciously high.

[SiriusHardware]

I just thought i'd give this a tentative nudge by uploading the .zip file you (Slothie) posted to this thread a while ago to JLCPCB, the PCB manufacturing outfit I think you said you have used before, just to see what the cost is at the moment.

https://jlcpcb.com/

When I hit 'Add Your Gerber' and upload the .zip file it seems to swallow it OK and fills in a lot of details for me, but when I go 'Add To Cart' it flags up the fact that I have not filled in the dimensions. I find this confusing, as I would have thought the dimensions were embedded in the design.

Do you remember having to supply the dimensions separately?

[Slothie]

No, one of the reasons I chose JLCPCB rather than PCBWAY was it got all the info from the zip file. Strange.

[Slothie]

The dimensions are on one of the drawing layers.

[SiriusHardware]

I've just tried again but using the file you attached to your original post about this in the 'vdu' thread - that one uploads OK, and including with dimensions.

Appx cost at present is about £12(!) for 5 PCBs and about the same again for postage.

[Slothie]

That's cheap! Does the preview look the same as it does in kicad? I uploaded the file to bitbucket and the forum from the same file.... Strange!

[SiriusHardware]

Yes, broadly, I mean all its arms and legs are attached. It looks OK in the JLCPCB preview.

I didn't go further yet because I would have had to create a sign-in / profile. I'll do that when I am feeling a bit more alert.

[Timbucus]

This is sounding promising...