Archive of the former Yahoo!Groups mailing list: Homebrew PCBs

Harvey,

I purposely didn't mention eyelets. Aside from the tedium of placing large numbers of them, the process to keep the flare on one side, while getting a proper flare on the other is quite tricky.  I have seen boards where the eyelets have been smashed flat.  While that seem to work, the solder connects the eyelet only at the periphery of the eyelet, leading to eventual failure with solder cracks at the periphery. Yes, with proper tooling, and good soldering practices they do work, but IMHO, they are not worth the trouble. They don't look good either :-)

Your lamination process sounds interesting.  I might try it st some point.  I am confused about the pins relative to the floor tiles.  How do you place the pins so that they don't interfere with the tiles?

(Another) Harvey

 

On Fri, 18 Mar 2016 20:21:11 -0700, you wrote:

>Brad,
>
>You are correct, this is quite a challenging project.
>
>Plated through holes:
>
>If you look at the board artwork in the article, you will find a view
>where they show the overlap between top and bottom patterns. At every
>location where the pads on the top and bottom have traces attached,
>there needs to be a connection between the top layer and the bottom
>layer. A board with plated through holes has a lining of copper plated
>into the holes to make those connections. Where there are no plated
>holes, the connection can be made by one of several methods. The
>easiest is to use the wires that go through the holes. For example,
>where a resistor or IC lead is in the hole, it would need to be soldered
>both top and bottom. For holes that just make a connection from the top
>to the bottom, a wire can be placed in the hole and soldered top and
>bottom. I usually make a small stub that is bent over at the end of the
>wire. I drop it into the hole and solder. Then I bend the wire the
>opposite way on the other side, and solder. This is called a "z"
>wire. Plated through holes are the best way to go, but are difficult to
>do in the home lab.

You can use eyelets, however, those are quite large in comparison, and
you need a source for them, and a staking tool.

>
>There are several complications associated with making two sided boards.
>The first, and most obvious, is that the two patterns have to be very
>closely aligned. This is required because of the connections between
>the two sizes. The holes must be drilled such that they pass through the
>pads on both sides. The actual etching process is also fraught with
>problems. You can choose to etch both sides at the same time, or
>protect one side from the etchant while etching the first side. Then
>reverse the procedure by protecting the first side and etching the
>second. The second process is tedious, but may result in better
>boards. It is hard to get both sides to etch evenly, and I frequently
>would wind up with over etching on one side when I tried to etch both at
>the same time. This is especially true with large boards such as the
>ones in this project.

I have tried both approaches, protecting a board side (you use
transparent contact paper), or not (you hope that it works well as you
have noted.)

However, I've done several things that seem to work well in
combination.

1) toner transfer top and bottom on two thinner boards. The material
I have is 0.021 or so and single sided, so that is good.

2) etch as needed.

3) drill the alignment holes ∗very∗ precisely. You'd like them within
about 0.002 of where they should be. I think I can get that close. I
can get rather thin vias to line up properly when using a #78 drill
(IIRC), using 10/10 for the process. I use #26 wire wrap wire,
stripped, for lacing the vias together.

However, I put a coating of long setting epoxy (about 1 hour work
time) on the boards. Cover the entire outside surface of the board
with masking tape, (keeps fingerprints off the copper), align the
boards with relatively thick pins, tape them so they cannot move, then
put them between two floor tiles to hold them flat while the epoxy
cures.

You get some well aligned boards that way, and if you screw up a
board, you only mess up one side.

Harvey

>
>(Another Harvey)
>
>
>On 3/18/2016 7:36 PM, 'Brad' unclefalter@... [Homebrew_PCBs] wrote:
>>
>> Separate discussion.
>>
>> I’m not planning to tackle this anytime soon, but I have been slowly
>> acquiring parts and am hoping after I’ve gained lots of experience to
>> one day build a Mark-8 replica.
>>
>> I expect this would be a very challenging project, which is why it’s
>> long in the future. But I’m always interested, always reading and
>> learning.
>>
>> Jon Titus, the creator, arranged with a New Jersey company to produce
>> board sets hobbyists could order, similar to the TVT. Those obviously
>> are ultra rare unobtanium now, however, I’m guessing the presence of
>> circuit board patterns in the construction details
>> (http://bytecollector.com/archive/mark_8/My_Mark-8_Info/Mark-8_Construction/RE_Mark-8_Const-NEW.pdf
>> -- page 25 or so) indicates it was at least possible some hobbyists
>> went and made their own, correct?
>>
>> I’m wondering – what differs in etching with double sided projects?
>> And when they say the holes weren’t plated through, does it mean what
>> I think it means (there would be a copper connection between a trace
>> on one side through a hole the component is soldered into through to
>> the other side?). Not sure I’m understanding how the connections
>> would be made from one side to the other.
>>
>>