Loading ...
Sorry, an error occurred while loading the content.

Re: [mill_drill] double check your drawings!

Expand Messages
  • MC Cason
    ... Now I understand the traces dead ending. I ve never dealt with anything that complicated. The last time I had to deal with multi layer PCBs was in the
    Message 1 of 20 , Jun 24, 2013
    • 0 Attachment
      On 06/24/2013 11:39 PM, Jerry Durand wrote:
      >
      > On 06/24/2013 09:24 PM, MC Cason wrote:
      > >
      > > The bad part is, that I programmed many of the components that I use,
      > > but I did not program the chips, somebody else had already done them. I
      > > now have every component that I use dimensioned correctly, and as close
      > > to looking like real world parts, as I could possibly make them.
      > >
      > > I have a few SMD components, and I have some 0603, and 0805 parts,
      > > which aren't too bad to work with. 0201 would probably drive me
      > > insane! I would have to upgrade all of my magnifiers to a vision
      > > system, to be able to work and test at that scale. Using BGA, I assume
      > > that you X-ray your boards?
      > >
      >
      > My wife does the PC board CAD. She doesn't worry about graphic pretty
      > of the parts, just accurate outlines and pin size/placement. That
      > particular board was 14 layers with I think 4 different drilling
      > operations (two of them laser).
      >

      Now I understand the traces dead ending. I've never dealt with
      anything that complicated. The last time I had to deal with multi layer
      PCBs was in the mid 90's. Looking at that board, it looks like I was
      lucky that I skipped the really extreme stuff.

      > The boards are x-rayed when we have to
      > use BGA for customers, but for our own smaller stuff (without BGA, just
      > QFN) we hand solder them here. 0402 is the smallest I'd like to hand
      > solder but I have done 0201 a few times. 0603 is our standard part.
      >

      I can deal with things as small as 0603, if I have my magnification
      where I can see clearly. Anything smaller is a pain.

      >
      > One trick my wife learned long ago, don't use ANY standard library
      > parts, there's almost always something wrong with them. She generally
      > makes up a new library for each new board, only pulling parts from her
      > other libraries if she can verify that they match current data sheets.
      >

      After that debacle, that's exactly what I do now. Not working in the
      electronics field in many years, caused a lot of rust to set in :)

      >
      > At least I'm no longer probing bare wafers under a 1500 power microscope.
      >

      I hated doing that. Luckily where I worked, we had a really nice
      vision system, that would clearly zoom in on a 30 mil area, and fill up
      a (10"?) CRT. That worked for most of what we did, but not for
      everything. I can still remember having to polish stripper feet under a
      microscope, because the scratches had to be incredibly small, and they
      could not cross completely across a vacuum port wall.

      One of the machines that I maintained, and the one that needed the
      most work, was a CNC sandblaster, that cut precision holes through a
      silicon wafer. We had a tight tolerance on holes, so it had to be
      adjusted constantly.

      --
      MC Cason
      Associate Developer - Eagle3D, Created by Matthias Weißer
    • Curt Wuollet
      You guys have it easy, lot s of the boards I did were done with tape and rubylith on a large scale then photoreduced. A cad screen is much easier on the eyes
      Message 2 of 20 , Jun 25, 2013
      • 0 Attachment
        You guys have it easy, lot's of the boards I did were done with tape and rubylith on a large scale then photoreduced. A cad screen is much easier on the eyes than working on a light table. And I had to take the ICs apart before probing them or doing voltage contrast on the SEM. CDC had their own little chip factory. I must say that cad has made it a lot easier and much less error prone making boards. It's nice to just submit gerbers and get boards back. It was a really big deal getting boards before.


        Regards

        cww

        On 06/25/2013 12:26 AM, MC Cason wrote:
         

        On 06/24/2013 11:39 PM, Jerry Durand wrote:
        >
        > On 06/24/2013 09:24 PM, MC Cason wrote:
        > >
        > > The bad part is, that I programmed many of the components that I use,
        > > but I did not program the chips, somebody else had already done them. I
        > > now have every component that I use dimensioned correctly, and as close
        > > to looking like real world parts, as I could possibly make them.
        > >
        > > I have a few SMD components, and I have some 0603, and 0805 parts,
        > > which aren't too bad to work with. 0201 would probably drive me
        > > insane! I would have to upgrade all of my magnifiers to a vision
        > > system, to be able to work and test at that scale. Using BGA, I assume
        > > that you X-ray your boards?
        > >
        >
        > My wife does the PC board CAD. She doesn't worry about graphic pretty
        > of the parts, just accurate outlines and pin size/placement. That
        > particular board was 14 layers with I think 4 different drilling
        > operations (two of them laser).
        >

        Now I understand the traces dead ending. I've never dealt with
        anything that complicated. The last time I had to deal with multi layer
        PCBs was in the mid 90's. Looking at that board, it looks like I was
        lucky that I skipped the really extreme stuff.

        > The boards are x-rayed when we have to
        > use BGA for customers, but for our own smaller stuff (without BGA, just
        > QFN) we hand solder them here. 0402 is the smallest I'd like to hand
        > solder but I have done 0201 a few times. 0603 is our standard part.
        >

        I can deal with things as small as 0603, if I have my magnification
        where I can see clearly. Anything smaller is a pain.

        >
        > One trick my wife learned long ago, don't use ANY standard library
        > parts, there's almost always something wrong with them. She generally
        > makes up a new library for each new board, only pulling parts from her
        > other libraries if she can verify that they match current data sheets.
        >

        After that debacle, that's exactly what I do now. Not working in the
        electronics field in many years, caused a lot of rust to set in :)

        >
        > At least I'm no longer probing bare wafers under a 1500 power microscope.
        >

        I hated doing that. Luckily where I worked, we had a really nice
        vision system, that would clearly zoom in on a 30 mil area, and fill up
        a (10"?) CRT. That worked for most of what we did, but not for
        everything. I can still remember having to polish stripper feet under a
        microscope, because the scratches had to be incredibly small, and they
        could not cross completely across a vacuum port wall.

        One of the machines that I maintained, and the one that needed the
        most work, was a CNC sandblaster, that cut precision holes through a
        silicon wafer. We had a tight tolerance on holes, so it had to be
        adjusted constantly.

        --
        MC Cason
        Associate Developer - Eagle3D, Created by Matthias Weißer


      • Jerry Durand
        ... I remember the mechanical marvel that was the ruby camera we had for making ICs, that was one big camera. Then we got a CAD system, took up two rooms and
        Message 3 of 20 , Jun 25, 2013
        • 0 Attachment
          On 06/25/2013 08:17 PM, Curt Wuollet wrote:
          You guys have it easy, lot's of the boards I did were done with tape and rubylith on a large scale then photoreduced. A cad screen is much easier on the eyes than working on a light table. And I had to take the ICs apart before probing them or doing voltage contrast on the SEM. CDC had their own little chip factory. I must say that cad has made it a lot easier and much less error prone making boards. It's nice to just submit gerbers and get boards back. It was a really big deal getting boards before.


          Regards

          cww

          I remember the mechanical marvel that was the ruby camera we had for making ICs, that was one big camera.

          Then we got a CAD system, took up two rooms and the pen plotter was another mechanical marvel.  I believe it was a 5 foot x 10 foot bed, used pressure-fed pens that were on an auto changer, and could take a day or two to make a plot.  The lead screws were the flap drivers off a Boeing aircraft and that thing could slide that I-beam pen holder from one end to the other amazingly fast.

          I got to design and wire up a rig so it could be left unattended, the operator could call in for status.  CalComp was impressed.  I also made a custom driver for the CAD system so it could use screens larger than a napkin.
          -- 
          Jerry Durand, Durand Interstellar, Inc.  www.interstellar.com
          tel: +1 408 356-3886, USA toll free: 1 866 356-3886
          Skype:  jerrydurand
        • philr_77378
          submit gerbers    What s that? Phil R ________________________________ From: Jerry Durand To: mill_drill@yahoogroups.com Sent:
          Message 4 of 20 , Jun 25, 2013
          • 0 Attachment
            "submit gerbers"   What's that?
            Phil R


            From: Jerry Durand <jdurand@...>
            To: mill_drill@yahoogroups.com
            Sent: Tuesday, June 25, 2013 10:26 PM
            Subject: Re: [mill_drill] double check your drawings!

             
            On 06/25/2013 08:17 PM, Curt Wuollet wrote:
            You guys have it easy, lot's of the boards I did were done with tape and rubylith on a large scale then photoreduced. A cad screen is much easier on the eyes than working on a light table. And I had to take the ICs apart before probing them or doing voltage contrast on the SEM. CDC had their own little chip factory. I must say that cad has made it a lot easier and much less error prone making boards. It's nice to just submit gerbers and get boards back. It was a really big deal getting boards before.


            Regards

            cww

            I remember the mechanical marvel that was the ruby camera we had for making ICs, that was one big camera.

            Then we got a CAD system, took up two rooms and the pen plotter was another mechanical marvel.  I believe it was a 5 foot x 10 foot bed, used pressure-fed pens that were on an auto changer, and could take a day or two to make a plot.  The lead screws were the flap drivers off a Boeing aircraft and that thing could slide that I-beam pen holder from one end to the other amazingly fast.

            I got to design and wire up a rig so it could be left unattended, the operator could call in for status.  CalComp was impressed.  I also made a custom driver for the CAD system so it could use screens larger than a napkin.
            -- 
            Jerry Durand, Durand Interstellar, Inc.  www.interstellar.com
            tel: +1 408 356-3886, USA toll free: 1 866 356-3886
            Skype:  jerrydurand


          • Jerry Durand
            g-code files like any CNC machine uses. In this case they operate a plotter to make a film negative that s used to etch the boards. ... -- Jerry Durand,
            Message 5 of 20 , Jun 25, 2013
            • 0 Attachment
              g-code files like any CNC machine uses.  In this case they operate a plotter to make a film negative that's used to etch the boards.

              On 06/25/2013 08:50 PM, philr_77378@... wrote:
              "submit gerbers"   What's that?
              Phil R


              -- 
              Jerry Durand, Durand Interstellar, Inc.  www.interstellar.com
              tel: +1 408 356-3886, USA toll free: 1 866 356-3886
              Skype:  jerrydurand
            • Vern
              I had to do mine with a etch-a-sketch, a sharpie and a pie plate full of ferric nitrate on a Bunsen burner. I had to build my ic s from discrete devices,
              Message 6 of 20 , Jun 25, 2013
              • 0 Attachment
                 I had to do mine with a etch-a-sketch, a sharpie and a pie plate full of ferric nitrate on a Bunsen burner.   I had to build my ic's from discrete devices, chicken toe nails and super glue...  Then test it all with a light bulb and a vacuum tube theremin.  So there.  ;)

                Sent from my iPhone

                On Jun 25, 2013, at 8:17 PM, Curt Wuollet <wideopen1@...> wrote:

                You guys have it easy, lot's of the boards I did were done with tape and rubylith on a large scale then photoreduced. A cad screen is much easier on the eyes than working on a light table. And I had to take the ICs apart before probing them or doing voltage contrast on the SEM. CDC had their own little chip factory. I must say that cad has made it a lot easier and much less error prone making boards. It's nice to just submit gerbers and get boards back. It was a really big deal getting boards before.


                Regards

                cww

                On 06/25/2013 12:26 AM, MC Cason wrote:
                 

                On 06/24/2013 11:39 PM, Jerry Durand wrote:
                >
                > On 06/24/2013 09:24 PM, MC Cason wrote:
                > >
                > > The bad part is, that I programmed many of the components that I use,
                > > but I did not program the chips, somebody else had already done them. I
                > > now have every component that I use dimensioned correctly, and as close
                > > to looking like real world parts, as I could possibly make them.
                > >
                > > I have a few SMD components, and I have some 0603, and 0805 parts,
                > > which aren't too bad to work with. 0201 would probably drive me
                > > insane! I would have to upgrade all of my magnifiers to a vision
                > > system, to be able to work and test at that scale. Using BGA, I assume
                > > that you X-ray your boards?
                > >
                >
                > My wife does the PC board CAD. She doesn't worry about graphic pretty
                > of the parts, just accurate outlines and pin size/placement. That
                > particular board was 14 layers with I think 4 different drilling
                > operations (two of them laser).
                >

                Now I understand the traces dead ending. I've never dealt with
                anything that complicated. The last time I had to deal with multi layer
                PCBs was in the mid 90's. Looking at that board, it looks like I was
                lucky that I skipped the really extreme stuff.

                > The boards are x-rayed when we have to
                > use BGA for customers, but for our own smaller stuff (without BGA, just
                > QFN) we hand solder them here. 0402 is the smallest I'd like to hand
                > solder but I have done 0201 a few times. 0603 is our standard part.
                >

                I can deal with things as small as 0603, if I have my magnification
                where I can see clearly. Anything smaller is a pain.

                >
                > One trick my wife learned long ago, don't use ANY standard library
                > parts, there's almost always something wrong with them. She generally
                > makes up a new library for each new board, only pulling parts from her
                > other libraries if she can verify that they match current data sheets.
                >

                After that debacle, that's exactly what I do now. Not working in the
                electronics field in many years, caused a lot of rust to set in :)

                >
                > At least I'm no longer probing bare wafers under a 1500 power microscope.
                >

                I hated doing that. Luckily where I worked, we had a really nice
                vision system, that would clearly zoom in on a 30 mil area, and fill up
                a (10"?) CRT. That worked for most of what we did, but not for
                everything. I can still remember having to polish stripper feet under a
                microscope, because the scratches had to be incredibly small, and they
                could not cross completely across a vacuum port wall.

                One of the machines that I maintained, and the one that needed the
                most work, was a CNC sandblaster, that cut precision holes through a
                silicon wafer. We had a tight tolerance on holes, so it had to be
                adjusted constantly.

                --
                MC Cason
                Associate Developer - Eagle3D, Created by Matthias Weißer


              • MC Cason
                ... Sounds like me, when I was in college. Pencil and paper, rub-on pads, a Sharpie, and hot Ferric Chloride in a glass dish, ICs were already fairly cheap,
                Message 7 of 20 , Jun 25, 2013
                • 0 Attachment
                  On 06/25/2013 11:24 PM, Vern wrote:
                  > I had to do mine with a etch-a-sketch, a sharpie and a pie plate full
                  > of ferric nitrate on a Bunsen burner. I had to build my ic's from
                  > discrete devices, chicken toe nails and super glue... Then test it
                  > all with a light bulb and a vacuum tube theremin. So there. ;)
                  >
                  > Sent from my iPhone
                  >

                  Sounds like me, when I was in college.

                  Pencil and paper, rub-on pads, a Sharpie, and hot Ferric Chloride in
                  a glass dish, ICs were already fairly cheap, 555's aplenty, as well as
                  logic gates. I never got into vacuum tubes, they were pretty much
                  history by then...

                  Designing on a Etch-a-Sketch... HMM, I'll have to try that some
                  time. It may freak out my eldest granddaughter though.

                  --
                  MC Cason
                  Associate Developer - Eagle3D, Created by Matthias Weißer
                • Vern
                  Oops, yes I mistyped... Ferric nitrate will etch silver... Not sure if it acts on copper. My high school electronics instructor was analog-oriented, and tubes
                  Message 8 of 20 , Jun 26, 2013
                  • 0 Attachment
                    Oops, yes I mistyped... Ferric nitrate will etch silver... Not sure if it acts on copper.

                    My high school electronics instructor was analog-oriented, and tubes were still available, though rapidly being phased out of many low power, consumer electronics. Even though the pocket calculator was all the rage, we were still taught to use a slide-rule. Sadly perhaps, I still know more about the physics of vacuum tubes ( or valves ) , than I do the bipolar NPN/PNP.

                    I admit the etch-a-sketch was a stretch... Might be worth the try, just to see her reaction. ;)

                    -V

                    Sent from my iPhone

                    On Jun 25, 2013, at 9:56 PM, MC Cason <farmerboy1967@...> wrote:

                    > On 06/25/2013 11:24 PM, Vern wrote:
                    >> I had to do mine with a etch-a-sketch, a sharpie and a pie plate full
                    >> of ferric nitrate on a Bunsen burner. I had to build my ic's from
                    >> discrete devices, chicken toe nails and super glue... Then test it
                    >> all with a light bulb and a vacuum tube theremin. So there. ;)
                    >>
                    >> Sent from my iPhone
                    >
                    > Sounds like me, when I was in college.
                    >
                    > Pencil and paper, rub-on pads, a Sharpie, and hot Ferric Chloride in
                    > a glass dish, ICs were already fairly cheap, 555's aplenty, as well as
                    > logic gates. I never got into vacuum tubes, they were pretty much
                    > history by then...
                    >
                    > Designing on a Etch-a-Sketch... HMM, I'll have to try that some
                    > time. It may freak out my eldest granddaughter though.
                    >
                    > --
                    > MC Cason
                    > Associate Developer - Eagle3D, Created by Matthias Weißer
                    >
                    >
                    >
                    >
                    > ------------------------------------
                    >
                    > Yahoo! Groups Links
                    >
                    >
                    >
                  • Curt Wuollet
                    Photoplotter files. At one time these drove an actual light plotter, now it s virtualized. Regards cww
                    Message 9 of 20 , Jun 26, 2013
                    • 0 Attachment
                      Photoplotter files. At one time these drove an actual light plotter, now it's virtualized.

                      Regards

                      cww


                      On 06/25/2013 10:50 PM, philr_77378@... wrote:
                       
                      "submit gerbers"   What's that?
                      Phil R


                      From: Jerry Durand <jdurand@...>
                      To: mill_drill@yahoogroups.com
                      Sent: Tuesday, June 25, 2013 10:26 PM
                      Subject: Re: [mill_drill] double check your drawings!

                       
                      On 06/25/2013 08:17 PM, Curt Wuollet wrote:
                      You guys have it easy, lot's of the boards I did were done with tape and rubylith on a large scale then photoreduced. A cad screen is much easier on the eyes than working on a light table. And I had to take the ICs apart before probing them or doing voltage contrast on the SEM. CDC had their own little chip factory. I must say that cad has made it a lot easier and much less error prone making boards. It's nice to just submit gerbers and get boards back. It was a really big deal getting boards before.


                      Regards

                      cww

                      I remember the mechanical marvel that was the ruby camera we had for making ICs, that was one big camera.

                      Then we got a CAD system, took up two rooms and the pen plotter was another mechanical marvel.  I believe it was a 5 foot x 10 foot bed, used pressure-fed pens that were on an auto changer, and could take a day or two to make a plot.  The lead screws were the flap drivers off a Boeing aircraft and that thing could slide that I-beam pen holder from one end to the other amazingly fast.

                      I got to design and wire up a rig so it could be left unattended, the operator could call in for status.  CalComp was impressed.  I also made a custom driver for the CAD system so it could use screens larger than a napkin.
                      -- 
                      Jerry Durand, Durand Interstellar, Inc.  www.interstellar.com
                      tel: +1 408 356-3886, USA toll free: 1 866 356-3886
                      Skype:  jerrydurand



                    • Curt Wuollet
                      They still design with tubes, there are very few 50,000 watt IC s. Regards cww ... They still design with tubes, there are very few 50,000 watt IC s. Regards
                      Message 10 of 20 , Jun 26, 2013
                      • 0 Attachment
                        They still design with tubes, there are very few 50,000 watt IC's.

                        Regards

                        cww


                        On 06/25/2013 11:56 PM, MC Cason wrote:
                         

                        On 06/25/2013 11:24 PM, Vern wrote:
                        > I had to do mine with a etch-a-sketch, a sharpie and a pie plate full
                        > of ferric nitrate on a Bunsen burner. I had to build my ic's from
                        > discrete devices, chicken toe nails and super glue... Then test it
                        > all with a light bulb and a vacuum tube theremin. So there. ;)
                        >
                        > Sent from my iPhone
                        >

                        Sounds like me, when I was in college.

                        Pencil and paper, rub-on pads, a Sharpie, and hot Ferric Chloride in
                        a glass dish, ICs were already fairly cheap, 555's aplenty, as well as
                        logic gates. I never got into vacuum tubes, they were pretty much
                        history by then...

                        Designing on a Etch-a-Sketch... HMM, I'll have to try that some
                        time. It may freak out my eldest granddaughter though.

                        --
                        MC Cason
                        Associate Developer - Eagle3D, Created by Matthias Weißer


                      • urrossum@att.net
                        ... And at the other end of the power spectrum, there are still some splinter areas that are enamored of tubes. I don t really get the tube hi-fi thing; it
                        Message 11 of 20 , Jun 27, 2013
                        • 0 Attachment
                          > They still design with tubes, there are very few 50,000 watt IC's.

                          And at the other end of the power spectrum, there are still some splinter areas that are enamored of tubes. I don't really "get" the tube hi-fi thing; it seems to me that anything other than precise, linear amplification (possibly with tone controls/equalization) is essentially distortion, regardless of how musical it may be.

                          However, in the actual music production context, that distortion can be quite useful. Enough so, in fact, that I'm now designing a couple of "boutique" tube pre-amps for the guitar and harmonica markets. I've learned a lot about the way tubes work over the last few weeks, and although I could probably model with reasonable accuracy the distortion of a starved-plate class-A stage using a DSP, it's just a lot easier to use the real thing. There's of course a certain cachet to doing it this way as well, and for the quantities I'm looking at there's still enough new-old-stock to make small production runs viable.

                          Don't get me started on the (non-)merits of "hand-wiring", though...
                          ~~
                          Mark Moulding
                        Your message has been successfully submitted and would be delivered to recipients shortly.