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

looking for a disk detect sensor

Expand Messages
  • B Degnan
    I am looking for a disk detect sensor for a PerSci 299 drive. Here is a picture with the part number (?).
    Message 1 of 12 , Mar 29, 2011
    • 0 Attachment
      I am looking for a disk detect sensor for a PerSci 299 drive. Here
      is a picture with the part number (?).

      http://www.vintagecomputer.net/persci/299revE/PerSci_299revE_Disk-Detect-Sensor.jpg

      (arm up)

      http://www.vintagecomputer.net/persci/299revE/PerSci_299revE_Disk-Detect-Sensor-Arm.jpg
      (arm down)

      The way this works is that a disk is inserted in the drive, pushing
      the arm into the sensor. The light is blocked and the sensor
      triggers the drive heads, etc to engage the disk.

      I tried various catalog sites, but I did not find anything close. I
      could use a hand to help locate a replacement part. From
      experimentation I don't think the little lamp inside of the sensor is
      the problem, it's the sensor itself.

      Bill
    • Barry L. Kline
      ... Hi Bill. Would something like this work? http://tinyurl.com/4bzjryc I don t know if the sensor in the picture is using an LED or is an older version with a
      Message 2 of 12 , Mar 30, 2011
      • 0 Attachment
        On 03/30/2011 01:11 AM, B Degnan wrote:
        > I am looking for a disk detect sensor for a PerSci 299 drive. Here
        > is a picture with the part number (?).

        Hi Bill.

        Would something like this work?

        http://tinyurl.com/4bzjryc

        I don't know if the sensor in the picture is using an LED or is an older
        version with a lamp. In either event I'm sure that you could easily
        modify the circuit to use this version.


        Barry
      • Mike Hatch
        ... That Mouser part OPB733TR is a surface mount item. Like Farnell and others, distributors typically use generic images that sometimes don t match the
        Message 3 of 12 , Mar 30, 2011
        • 0 Attachment
          --- In midatlanticretro@yahoogroups.com, "Barry L. Kline" <blkline@...> wrote:
          >
          > Would something like this work?
          >
          > http://tinyurl.com/4bzjryc
          >
          > I don't know if the sensor in the picture is using an LED or is an older
          > version with a lamp. In either event I'm sure that you could easily
          > modify the circuit to use this version.
          >
          >
          > Barry
          >
          That Mouser part OPB733TR is a surface mount item. Like Farnell and others, distributors typically use "generic" images that sometimes don't match the component - "Images are for reference only See Product Specifications"

          Bill's part -H13B1- is a LED-Darlington sensor in a very common package, newer types would work. In fixing old kit its the physical size first then the electrical properties as the replacement usually out-performs the old.

          An OPB360 / 370 series in package type T is very similar in size and the pin out is the same, slot width and device height is also very close. Optek is the manufacturer, OPB360T51 is stocked
          by Digikey in the US at $1.84, RS or Farnell in the UK at about £2.

          Best regards
          Mike
          www.soemtron.org
        • Mr Ian Primus
          ... That little black optical gate with the GE logo? Those are very common little opto-sensors. It s just a photodiode and an IR LED in a little black plastic
          Message 4 of 12 , Mar 30, 2011
          • 0 Attachment
            --- On Wed, 3/30/11, B Degnan <billdeg@...> wrote:

            > The way this works is that a disk is inserted in the drive,
            > pushing
            > the arm into the sensor. The light is blocked and the
            > sensor
            > triggers the drive heads, etc to engage the disk.
            >

            That little black optical gate with the GE logo? Those are very common little opto-sensors. It's just a photodiode and an IR LED in a little black plastic thing. Most of them come apart if you desolder them - the black shroud comes off revealing the two elements. Some are epoxy potted though. If it is the kind that comes apart, you can replace just the bad element. Photodiodes can usually be robbed from the index sensors on PC 5 1/4" drives.

            These are very interchangeable for most other parts that look like that. They're used in lots of things - printers, disk drives, video games, etc. The same thing is used in the joystick in GORF (Bally/Midway arcade game). I've replaced them in the past with parts robbed from printers, etc. I've never paid attention to the part numbers when replacing, it doesn't seem to matter.

            Alternatively, a quick google turned up one for five bucks on Amazon, if you want the exact part.

            http://www.amazon.com/American-Microsemiconductor-H13B1/dp/B003UQUQIC

            -Ian
          • Dan Roganti
            ~http://www.rogtronics.net ... yes, that part can work. Have you checked if the rest of that circuit board is functioning ? From
            Message 5 of 12 , Mar 30, 2011
            • 0 Attachment

              ~http://www.rogtronics.net


              On Wed, Mar 30, 2011 at 8:48 AM, Mike Hatch <mike@...> wrote:

              An OPB360 / 370 series in package type T is very similar in size and the pin out is the same, slot width and device height is also very close. Optek is the manufacturer, OPB360T51 is stocked
              by Digikey in the US at $1.84, RS or Farnell in the UK at about £2.


              yes, that part can work.
              Have you checked if the rest of that circuit board is functioning ?
              From looking at the manual page 2-41, you like to check if the logic circuit on there is still working.
              You can do so by using a logic probe and jumpering a bypass at the connection of the opto on the transistor side, at the input of U4 schmitt inverter - when looking at the schematic on the manual - by touching the C and E of the opto together - in effect simulating the opto. Then probing with a logic probe to see if it changes from Lo to Hi anf following thru to the output pin on the connector for that little board.

              =Dan

            • B. Degnan
              Thanks Mike, Barry, Ian, and Dan. When I see pages like this http://www.amazon.com/American-Microsemiconductor-H13B1/dp/B003UQUQIC ...where the picture is so
              Message 6 of 12 , Mar 30, 2011
              • 0 Attachment
                Thanks Mike, Barry, Ian, and Dan.

                When I see pages like this
                http://www.amazon.com/American-Microsemiconductor-H13B1/dp/B003UQUQIC
                ...where the picture is so way off to the item it makes me think I have the wrong thing, but I see that the description is "H13B1 Photo Gap Detector" so it must be correct, thanks.

                In some photos I was holding up/back the arm to get a better picture. It does go all of the way in.

                I believe the problem is with the sensor itself (or circuit) and not the lamp being burned out. What I did to prove this was shine a UV light at the sensor to fool it into thinking there is a disk in the drive. Drive 1 will open/close in response to the light, but drive 0 stays still. If the light were bad my experiment would cause the sensor to trip.

                >
                > >
                > > An OPB360 / 370 series in package type T is very similar in size and the
                > > pin out is the same, slot width and device height is also very close. Optek
                > > is the manufacturer, OPB360T51 is stocked
                > > by Digikey in the US at $1.84, RS or Farnell in the UK at about £2.
                > >
                >
                >
                > yes, that part can work.
                > Have you checked if the rest of that circuit board is functioning ?

                No. It's tough because the drive is inserted into the drive bay and I have to feel around to attach the power and data leads. I'd have to totally take apart the entire desk to get serious about testing. FYI Drive 1 works. I can even boot the computer from drive 1, but the OS is not BIOSed to use drive 1 as the primary drive. After the boot the system tries to use drive 0 and hangs. The cards all work, the disks are OK. I am confident that I have narrowed the problem down to the physical sensor, or ajoining compontent/circuit.


                > From looking at the manual page 2-41, you like to check if the logic circuit
                > on there is still working.

                What rev drive's manual? The manual that came with the replacement drive I bought is for a later version of the PerSci 299 (rev H), totally different than the older version 299 that came with the computer (although the name plate on both says "299"). The older rev E 299 uses an 8080 as the brains of the drive, kind of like how a 6502 controls a CBM 1541 drive. The rev H version detailed in the manual has only TTL logic and no CPU.

                > You can do so by using a logic probe and jumpering a bypass at the
                > connection of the opto on the transistor side, at the input of U4 schmitt
                > inverter - when looking at the schematic on the manual - by touching the C
                > and E of the opto together - in effect simulating the opto. Then probing
                > with a logic probe to see if it changes from Lo to Hi anf following thru to
                > the output pin on the connector for that little board.
                >

                It's relatively easy to put in a new sensor, although I have to remove the entire drive engage mechanism to do get to it. If replacing the sensor does not work, then it will be time to take the computer and drive out of the desk and get serious with it.

                Bill
              • Dan Roganti
                ~{sent *[from ±my] ^Android} ... lamp being burned out. What I did to prove this was shine a UV light at the sensor to fool it into thinking there is a disk
                Message 7 of 12 , Mar 30, 2011
                • 0 Attachment

                  ~{sent *[from ±my] ^Android}

                  On Mar 30, 2011 9:50 AM, "B. Degnan" <billdeg@...> wrote:
                  >
                  > I believe the problem is with the sensor itself (or circuit) and not the lamp being burned out.  What I did to prove this was shine a UV light at the sensor to fool it into thinking there is a disk in the drive.  

                  You really want IR
                  UV is at the other end of the spectrum :)
                  As in a TV remote
                  You can usethe logic probe there to and check it the opto ouput changes. Notice in the manual the output is on the emitter.

                  Drive 1 will open/close in response to the light, but drive 0 stays still.  If the light were bad my experiment would cause the sensor to trip.
                  >
                  > >
                  > > >

                • B. Degnan
                  ... the ... the ... Yes, but it works! I think it s just that it s a very bright grow light bulb that I was using to get a better view of what was going on
                  Message 8 of 12 , Mar 30, 2011
                  • 0 Attachment
                    > >
                    > > I believe the problem is with the sensor itself (or circuit) and not
                    the
                    > lamp being burned out. What I did to prove this was shine a UV light at
                    the
                    > sensor to fool it into thinking there is a disk in the drive.
                    >
                    > You really want IR
                    > UV is at the other end of the spectrum :)
                    > As in a TV remote

                    Yes, but it works! I think it's just that it's a very bright "grow light"
                    bulb that I was using to get a better view of what was going on in there.
                    I did not expect that this light would trip the sensor. A regular
                    florescent does not work.


                    > You can usethe logic probe there to and check it the opto ouput changes.
                    > Notice in the manual the output is on the emitter.
                    >
                    > Drive 1 will open/close in response to the light, but drive 0 stays
                    still.
                    > If the light were bad my experiment would cause the sensor to trip.
                    > >
                    > > >
                    > > > >
                  • Dan Roganti
                    ... I missed that part :) I have to quit using my smartphone to read email
                    Message 9 of 12 , Mar 30, 2011
                    • 0 Attachment

                      On Mar 30, 2011 10:38 AM, "B. Degnan" <billdeg@...> wrote:
                      > Yes, but it works!  I think it's just that it's a very bright "grow light"
                      > bulb that I was using to get a better view of what was going on in there.
                      > I did not expect that this light would trip the sensor.  A regular
                      > florescent does not work.
                      >

                      I missed that part :)
                      I have to quit using my smartphone to read email

                    • systems_glitch
                      ... Yep, they re usually IR detectors. I ve replaced the source in the source-sensor pair with a regular IR LED, just by drilling a hole through the back of
                      Message 10 of 12 , Mar 30, 2011
                      • 0 Attachment

                        > Yes, but it works!  I think it's just that it's a very bright "grow light"
                        > bulb that I was using to get a better view of what was going on in there.
                        > I did not expect that this light would trip the sensor.  A regular
                        > florescent does not work.

                        Yep, they're usually IR detectors. I've replaced the source in the source-sensor pair with a regular IR LED, just by drilling a hole through the back of the pair. Usually I've had to adjust the gain of the input amp.
                         
                        You can also check to see if the source in the pair is working by aiming a CCD camera at it. Cheap webcams are the best, as they have no IR filters.
                         
                        Jonathan
                      • B. Degnan
                        ... there. ... through ... amp. ... a ... filters. ... Maybe a grow lignt sends both IR and UltraViolet? This is reasonable to believe. This would also
                        Message 11 of 12 , Mar 30, 2011
                        • 0 Attachment
                          > >
                          > > > Yes, but it works! I think it's just that it's a very bright "grow
                          > > light"
                          > > > bulb that I was using to get a better view of what was going on in
                          there.
                          > > > I did not expect that this light would trip the sensor. A regular
                          > > > florescent does not work.
                          > >
                          > Yep, they're usually IR detectors. I've replaced the source in the
                          > source-sensor pair with a regular IR LED, just by drilling a hole
                          through
                          > the back of the pair. Usually I've had to adjust the gain of the input
                          amp.
                          >
                          > You can also check to see if the source in the pair is working by aiming
                          a
                          > CCD camera at it. Cheap webcams are the best, as they have no IR
                          filters.
                          >
                          > Jonathan

                          Maybe a grow lignt sends both IR and UltraViolet? This is reasonable to
                          believe. This would also explain why a plain flourescent light did not
                          trigger the sensor. eh?

                          anyway, I ordered the part from Amazon per Ian's link. It should take only
                          45 minutes to install.

                          bd
                        • Dan Roganti
                          ... what s with the grow light ;) you have to stop mixing your paraphernalia with your peripherals ~haa
                          Message 12 of 12 , Mar 30, 2011
                          • 0 Attachment
                            On Wed, Mar 30, 2011 at 4:27 PM, B. Degnan <billdeg@...> wrote:
                            >

                            Maybe a grow lignt sends both IR and UltraViolet?  This is reasonable to
                            believe.  This would also explain why a plain flourescent light did not
                            trigger the sensor.  eh?

                            anyway, I ordered the part from Amazon per Ian's link.  It should take only
                            45 minutes to install.


                            what's with the grow light ;)
                            you have to stop mixing your paraphernalia with your peripherals ~haa

                          Your message has been successfully submitted and would be delivered to recipients shortly.