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

Re: [softrock40] Si570 desoldering tips

Expand Messages
  • Dave - WB6DHW
    ... Hot air. Just be careful not to overheat nearby components. Dave - WB6DHW
    Message 1 of 20 , Feb 1, 2010
    • 0 Attachment
      barrymichels wrote:
      > I just received a shipment notification for the 2 Si570 samples I ordered 2 weeks ago. They should be here Wednesday. The Si570 that is in my RXTX won't let me tune over 10Mhz. What's the best method for removing the damaged chip without damaging the board?
      >
      > Thanks,
      > Barry - KE4JUC
      >
      >
      >
      >
      Hot air. Just be careful not to overheat nearby components.


      Dave - WB6DHW
      <http://wb6dhw.com>
    • robnj74@aol.com
      Hot Air as previously suggested is IMHO the easiest way, I normally shield the nearby components with tin-foil. It can prevent damage to the nearby
      Message 2 of 20 , Feb 1, 2010
      • 0 Attachment
        Hot Air as previously suggested is IMHO the easiest way, I normally shield the nearby components with  tin-foil. It can prevent damage to the nearby components.
         
        73's Rob K2RJJ
         
        In a message dated 2/1/2010 8:43:42 P.M. Eastern Standard Time, brainerd@... writes:
         

        barrymichels wrote:
        > I just received a shipment notification for the 2 Si570 samples I ordered 2 weeks ago. They should be here Wednesday. The Si570 that is in my RXTX won't let me tune over 10Mhz. What's the best method for removing the damaged chip without damaging the board?
        >
        > Thanks,
        > Barry - KE4JUC
        >
        >
        >
        >
        Hot air. Just be careful not to overheat nearby components.

        Dave - WB6DHW
        <http://wb6dhw. com>

      • g3oth
        Hi Barry Back in December 31 2008 I started a thread called Si570 Removal Tips Message Number 29377. Go to the top of this list and put that message number in
        Message 3 of 20 , Feb 1, 2010
        • 0 Attachment
          Hi Barry
          Back in December 31 2008 I started a thread called Si570 Removal Tips Message Number 29377.
          Go to the top of this list and put that message number in the box and take a look at what I did and what others had to say back then.
          Best of luck with the job, you will find it is quite easy to do.
          73 de Charles G3OTH


          --- In softrock40@yahoogroups.com, "barrymichels" <bmichels@...> wrote:
          >
          > I just received a shipment notification for the 2 Si570 samples I ordered 2 weeks ago. They should be here Wednesday. The Si570 that is in my RXTX won't let me tune over 10Mhz. What's the best method for removing the damaged chip without damaging the board?
          >
          > Thanks,
          > Barry - KE4JUC
          >
        • Alan
          ... From: barrymichels Subject: [softrock40] Si570 desoldering tips ... Are you sure of this? It is just possible something else could be giving this effect.
          Message 4 of 20 , Feb 1, 2010
          • 0 Attachment
            ----- Original Message -----
            From: "barrymichels"
            Subject: [softrock40] Si570 desoldering tips


            > The Si570 that is in my RXTX won't let me tune over 10Mhz.

            Are you sure of this? It is just possible something else could be giving
            this effect.

            73 Alan G4ZFQ
          • Barry Michels
            ... Well, since accidentally frying my RXTX, I ve replaced every piece of silicon on it. The only thing left is the Si570. I have noise bands all over the
            Message 5 of 20 , Feb 2, 2010
            • 0 Attachment
              Alan wrote:
              > ----- Original Message -----
              > From: "barrymichels"
              > Subject: [softrock40] Si570 desoldering tips
              >
              >
              >
              >> The Si570 that is in my RXTX won't let me tune over 10Mhz.
              >>
              >
              > Are you sure of this? It is just possible something else could be giving
              > this effect.
              >
              > 73 Alan G4ZFQ
              >
              >
              Well, since accidentally frying my RXTX, I've replaced every piece of
              silicon on it. The only thing left is the Si570. I have noise bands
              all over the waterfall that are almost gone at 80M, but get worse as you
              go up in frequency. I can receive WWV at 10Mhz, but no higher. Above
              that, I get no signals at all. Just a very large/wide center spike and
              noise bands on the edges. I've been running WSPR on 40M for weeks now
              with no trouble, but am missing out on 30M and 20M activity. I'm hoping
              the new Si570 will fix the problem. If not, I bought enough spare chips
              to do a full replacement again (except for the TX OpAmp, but that's
              working fine).


              Barry - KE4JUC
            • Alan
              ... From: Barry Michels Subject: Re: [softrock40] Si570 desoldering tips ... You have some problems! I just wondered if you had checked the output of the
              Message 6 of 20 , Feb 2, 2010
              • 0 Attachment
                ----- Original Message -----
                From: "Barry Michels"
                Subject: Re: [softrock40] Si570 desoldering tips


                >>
                > Well, since accidentally frying my RXTX, I've replaced every piece of
                > silicon on it. The only thing left is the Si570. I have noise bands
                > all over the waterfall that are almost gone at 80M,


                You have some problems! I just wondered if you had checked the output of the
                Si570.
                You are not likely to damage it by following the advice, as you say, there
                does not seem to be much else left!

                73 Alan G4ZFQ .
              • Mr Doug -
                In a situation like yours, where the entire board is fried, it would seem a better approach would be to just assemble a new board and throw that board in the
                Message 7 of 20 , Feb 2, 2010
                • 0 Attachment
                  In a situation like yours, where the entire board is fried, it would seem a better approach would be to just assemble a new board and throw that board in the trash, minus whatever you want to save from it. That is unless you like the experience and possible aggravation of fixing it! The cost of new parts, the possible destruction of the board, your time, and the fact that these boards are VERY inexpensively priced yo begin with would seem to make them throwaways in some situations. Granted you might be beyond that now but something to consider in the future.



                  From: Barry Michels <bmichels@...>
                  To: softrock40@yahoogroups.com
                  Sent: Tue, February 2, 2010 8:30:53 AM
                  Subject: Re: [softrock40] Si570 desoldering tips

                   

                  Alan wrote:

                  > ----- Original Message -----
                  > From: "barrymichels"
                  > Subject: [softrock40] Si570 desoldering tips
                  >
                  >
                  >
                  >> The Si570 that is in my RXTX won't let me tune over 10Mhz.
                  >>
                  >
                  > Are you sure of this? It is just possible something else could be giving
                  > this effect.
                  >
                  > 73 Alan G4ZFQ
                  >
                  >
                  Well, since accidentally frying my RXTX, I've replaced every piece of
                  silicon on it. The only thing left is the Si570. I have noise bands
                  all over the waterfall that are almost gone at 80M, but get worse as you
                  go up in frequency. I can receive WWV at 10Mhz, but no higher. Above
                  that, I get no signals at all. Just a very large/wide center spike and
                  noise bands on the edges. I've been running WSPR on 40M for weeks now
                  with no trouble, but am missing out on 30M and 20M activity. I'm hoping
                  the new Si570 will fix the problem. If not, I bought enough spare chips
                  to do a full replacement again (except for the TX OpAmp, but that's
                  working fine).

                  Barry - KE4JUC


                • Barry Michels
                  So far, I ve only got about $25 in repair costs. One Mouser order and a small order from Tony, plus my free Si570 samples. I don t mind the time spent. It
                  Message 8 of 20 , Feb 2, 2010
                  • 0 Attachment
                    So far, I've only got about $25 in repair costs. One Mouser order and a
                    small order from Tony, plus my free Si570 samples. I don't mind the
                    time spent. It took about 7 hours to build the original (including a
                    PA/F kit and all 4 filter modules). Don't really want to go thru it
                    again from scratch.


                    Barry

                    Mr Doug - wrote:
                    >
                    >
                    > In a situation like yours, where the entire board is fried, it would
                    > seem a better approach would be to just assemble a new board and throw
                    > that board in the trash, minus whatever you want to save from it. That
                    > is unless you like the experience and possible aggravation of fixing
                    > it! The cost of new parts, the possible destruction of the board, your
                    > time, and the fact that these boards are VERY inexpensively priced yo
                    > begin with would seem to make them throwaways in some situations.
                    > Granted you might be beyond that now but something to consider in the
                    > future.
                    >
                    >
                    > ------------------------------------------------------------------------
                    > *From:* Barry Michels <bmichels@...>
                    > *To:* softrock40@yahoogroups.com
                    > *Sent:* Tue, February 2, 2010 8:30:53 AM
                    > *Subject:* Re: [softrock40] Si570 desoldering tips
                    >
                    >
                    >
                    > Alan wrote:
                    > > ----- Original Message -----
                    > > From: "barrymichels"
                    > > Subject: [softrock40] Si570 desoldering tips
                    > >
                    > >
                    > >
                    > >> The Si570 that is in my RXTX won't let me tune over 10Mhz.
                    > >>
                    > >
                    > > Are you sure of this? It is just possible something else could be
                    > giving
                    > > this effect.
                    > >
                    > > 73 Alan G4ZFQ
                    > >
                    > >
                    > Well, since accidentally frying my RXTX, I've replaced every piece of
                    > silicon on it. The only thing left is the Si570. I have noise bands
                    > all over the waterfall that are almost gone at 80M, but get worse as you
                    > go up in frequency. I can receive WWV at 10Mhz, but no higher. Above
                    > that, I get no signals at all. Just a very large/wide center spike and
                    > noise bands on the edges. I've been running WSPR on 40M for weeks now
                    > with no trouble, but am missing out on 30M and 20M activity. I'm hoping
                    > the new Si570 will fix the problem. If not, I bought enough spare chips
                    > to do a full replacement again (except for the TX OpAmp, but that's
                    > working fine).
                    >
                    > Barry - KE4JUC
                    >
                    >
                    >
                    >
                    >
                    >
                    > --
                    > This message has been scanned for viruses and
                    > dangerous content by *MailScanner* <http://www.mailscanner.info/>, and is
                    > believed to be clean.
                  • Preston Douglas
                    Barry, Could you please tell us how you fried your RXTX board? While I agree with Doug that the boards are inexpensive to replace, the sweat equity invested
                    Message 9 of 20 , Feb 2, 2010
                    • 0 Attachment
                      Barry,

                      Could you please tell us how you fried your RXTX board?  While I agree with Doug that the boards are inexpensive to replace, the sweat equity invested in one is considerable.  I am happy my RXTX 6.3 works well, and would not want to build another one all over again.  So, how did you kill yours, so we can avoid doing whatever it is you did!


                      Preston Douglas WJ2V
                    • Barry Michels
                      I m not sure how exactly. It s not in a case. I am guessing that maybe 12V went into the antenna input. This was before installing any PA/F module and I had
                      Message 10 of 20 , Feb 2, 2010
                      • 0 Attachment
                        I'm not sure how exactly. It's not in a case. I am guessing that maybe
                        12V went into the antenna input. This was before installing any PA/F
                        module and I had the antenna connector jumpered to bypass that module.
                        One day it just stopped receiving anything but a strong local AM station
                        in the center and 2 other places on the waterfall. Q2 failed closed
                        (stuck in transmit).

                        Lesson: put your RXTX in a case of any kind after confirming that it
                        works. Even a plastic tupperware container and don't let any metal
                        brush by it.

                        Barry - KE4JUC

                        Preston Douglas wrote:
                        >
                        >
                        > Barry,
                        >
                        > Could you please tell us how you fried your RXTX board? While I agree
                        > with Doug that the boards are inexpensive to replace, the sweat equity
                        > invested in one is considerable. I am happy my RXTX 6.3 works well,
                        > and would not want to build another one all over again. So, how did
                        > you kill yours, so we can avoid doing whatever it is you did!
                        >
                        >
                        > Preston Douglas WJ2V
                        >
                        >
                        >
                        >
                        > --
                        > This message has been scanned for viruses and
                        > dangerous content by *MailScanner* <http://www.mailscanner.info/>, and is
                        > believed to be clean.
                      • Barry Michels
                        ... Just replaced the Si570 and am sad to report that nothing has changed. I learned something in the process, though. After reading all the suggestions and
                        Message 11 of 20 , Feb 3, 2010
                        • 0 Attachment
                          Barry Michels wrote:
                          > Alan wrote:
                          >
                          >> ----- Original Message -----
                          >> From: "barrymichels"
                          >> Subject: [softrock40] Si570 desoldering tips
                          >>
                          >>
                          >>
                          >>
                          >>> The Si570 that is in my RXTX won't let me tune over 10Mhz.
                          >>>
                          >>>
                          >> Are you sure of this? It is just possible something else could be giving
                          >> this effect.
                          >>
                          >> 73 Alan G4ZFQ
                          >>
                          >>
                          >>
                          > Well, since accidentally frying my RXTX, I've replaced every piece of
                          > silicon on it. The only thing left is the Si570. I have noise bands
                          > all over the waterfall that are almost gone at 80M, but get worse as you
                          > go up in frequency. I can receive WWV at 10Mhz, but no higher. Above
                          > that, I get no signals at all. Just a very large/wide center spike and
                          > noise bands on the edges. I've been running WSPR on 40M for weeks now
                          > with no trouble, but am missing out on 30M and 20M activity. I'm hoping
                          > the new Si570 will fix the problem. If not, I bought enough spare chips
                          > to do a full replacement again (except for the TX OpAmp, but that's
                          > working fine).
                          >
                          >
                          > Barry - KE4JUC
                          >
                          >

                          Just replaced the Si570 and am sad to report that nothing has changed.
                          I learned something in the process, though. After reading all the
                          suggestions and looking around my house, I ended up desoldering the
                          Si570 by folding a piece of heavy duty aluminum foil so I had 3
                          thicknesses and cut out a square just the size of the chip. Laid the
                          aluminum foil over the board and wrapped the edges around my board vise
                          to hold it in place. Put slight pressure on one side of the chip with a
                          long screwdriver and heated the chip from a distance with my wife's
                          creme brulee torch. Bringing the flame closer to the board, the chip
                          popped off and landed on the foil with no damage done to the traces
                          underneath. Only slight heat marks on the top layer of foil. Pretty
                          painless and no trip to the hobby shop needed.

                          I replaced the RX mixer with no improvement. Not sure where to go from
                          here. I'd hate to scrap it. If nothing else, I'll just keep this unit
                          for a 80/40 QRP radio and build another when the Mobo or widget projects
                          are a little more mature...

                          I've uploaded screen captures of Rocky tuned above 10Mhz. Look in the
                          photos section under KE4JUC.


                          Barry - KE4JUC
                        • g8voip
                          Hi Barry, Looking at your Rocky screen grab, you have no image rejection at all. Although that would not explain your apparent tuning problem above 10MHz,
                          Message 12 of 20 , Feb 3, 2010
                          • 0 Attachment
                            Hi Barry,

                            Looking at your Rocky screen grab, you have no image rejection at all. Although that would not explain your apparent tuning problem above 10MHz, there is something pretty fundamental wrong with your receiver or PC / soundcard set up / configuration.

                            I think you need to go back to basics and determine what is actually still working on your PCB.

                            A good place to start is to find out exactly what the Si570 LO is doing.

                            Do you have a frequency counter available you could hook up on say U9 pin 3 (clock into the 74HC74. If you use Fred PE0FKO's 'CFGSR' program, you can use the 'Tune' tab and select 1MHz steps and easily tune from 1MHz up to 30MHz, and you should be able to measure the Si570 output which will be at 4x the 'VFO' frequency ( 4 - 120MHz). That will confirm whether or not your LO is functioning correctly.

                            If that is ok, then you will need to go on to the divider stage and confirm the two outputs to the RX mixer are correct. Access to an oscilloscope would be useful to see what is happening.

                            In effect, follow the original build / checkout procedure.

                            73, Bob G8VOI
                          • Barry Michels
                            ... When I get some more time, I will run thru the checkout procedures. Below 10Mhz, I get very flat response (no hump in the middle, just a thin spike) with
                            Message 13 of 20 , Feb 3, 2010
                            • 0 Attachment
                              g8voip wrote:
                              > Hi Barry,
                              >
                              > Looking at your Rocky screen grab, you have no image rejection at all. Although that would not explain your apparent tuning problem above 10MHz, there is something pretty fundamental wrong with your receiver or PC / soundcard set up / configuration.
                              >
                              > I think you need to go back to basics and determine what is actually still working on your PCB.
                              >
                              > A good place to start is to find out exactly what the Si570 LO is doing.
                              >
                              > Do you have a frequency counter available you could hook up on say U9 pin 3 (clock into the 74HC74. If you use Fred PE0FKO's 'CFGSR' program, you can use the 'Tune' tab and select 1MHz steps and easily tune from 1MHz up to 30MHz, and you should be able to measure the Si570 output which will be at 4x the 'VFO' frequency ( 4 - 120MHz). That will confirm whether or not your LO is functioning correctly.
                              >
                              > If that is ok, then you will need to go on to the divider stage and confirm the two outputs to the RX mixer are correct. Access to an oscilloscope would be useful to see what is happening.
                              >
                              > In effect, follow the original build / checkout procedure.
                              >
                              > 73, Bob G8VOI
                              >
                              >
                              When I get some more time, I will run thru the checkout procedures.

                              Below 10Mhz, I get very flat response (no hump in the middle, just a
                              thin spike) with great image rejection. The screenshots I posted only
                              look like that above 10Mhz (more like 10.2Mhz). Last time I checked, I
                              could hear the LO output on a nearby receiver even above 10Mhz. Will
                              try that again later.

                              The only oscilloscope I have access to is my 32Mhz digital scope that
                              isn't worth much other than working with digital signals. It's an old
                              PCS32 parallel port model from Velleman.


                              Thanks for the help!
                              Barry - KE4JUC
                            • g8voip
                              Hi Barry, An obvious question, but you are actually changing over and using the different RX filter and TX PA / filter modules when you are trying to receive
                              Message 14 of 20 , Feb 3, 2010
                              • 0 Attachment
                                Hi Barry,

                                An obvious question, but you are actually changing over and using the different RX filter and TX PA / filter modules when you are trying to receive on the higher frequency bands?

                                To have good image rejection at the lower frequency suggests a problem with the phase splitter transformer on the RX filter module for the higher band.

                                If you are listening for the LO on another receiver, remember the LO is running at 4 x the tuned frequency.

                                To me it looks like you might have a combination of problems.

                                You never did comment quite how you managed to destroy your v6.3 TX/RX in the first place. Knowing that might provide some clues to what has gone wrong, assuming that the board was completely working on all bands before the problem occurred.

                                73, Bob G8VOI




                                --- In softrock40@yahoogroups.com, Barry Michels <bmichels@...> wrote:
                                >
                                > g8voip wrote:
                                > > Hi Barry,
                                > >
                                > > Looking at your Rocky screen grab, you have no image rejection at all. Although that would not explain your apparent tuning problem above 10MHz, there is something pretty fundamental wrong with your receiver or PC / soundcard set up / configuration.
                                > >
                                > > I think you need to go back to basics and determine what is actually still working on your PCB.
                                > >
                                > > A good place to start is to find out exactly what the Si570 LO is doing.
                                > >
                                > > Do you have a frequency counter available you could hook up on say U9 pin 3 (clock into the 74HC74. If you use Fred PE0FKO's 'CFGSR' program, you can use the 'Tune' tab and select 1MHz steps and easily tune from 1MHz up to 30MHz, and you should be able to measure the Si570 output which will be at 4x the 'VFO' frequency ( 4 - 120MHz). That will confirm whether or not your LO is functioning correctly.
                                > >
                                > > If that is ok, then you will need to go on to the divider stage and confirm the two outputs to the RX mixer are correct. Access to an oscilloscope would be useful to see what is happening.
                                > >
                                > > In effect, follow the original build / checkout procedure.
                                > >
                                > > 73, Bob G8VOI
                                > >
                                > >
                                > When I get some more time, I will run thru the checkout procedures.
                                >
                                > Below 10Mhz, I get very flat response (no hump in the middle, just a
                                > thin spike) with great image rejection. The screenshots I posted only
                                > look like that above 10Mhz (more like 10.2Mhz). Last time I checked, I
                                > could hear the LO output on a nearby receiver even above 10Mhz. Will
                                > try that again later.
                                >
                                > The only oscilloscope I have access to is my 32Mhz digital scope that
                                > isn't worth much other than working with digital signals. It's an old
                                > PCS32 parallel port model from Velleman.
                                >
                                >
                                > Thanks for the help!
                                > Barry - KE4JUC
                                >
                              • Preston Douglas
                                Y know, Barry, something you mentioned has been running around in my head (along with the squirrels!). If the way you killed your 6.3 was by misconnecting the
                                Message 15 of 20 , Feb 3, 2010
                                • 0 Attachment
                                  Y'know, Barry, something you mentioned has been running around in my head (along with the squirrels!).  If the way you killed your 6.3 was by misconnecting the PA daughterboard so you shorted 12V to the amplifier board, you can't burn out chips that way.  They are isolated from the antenna inputs.  All you can do is burn out the secondary of the output transformer or the output inductor in the little RX filter.  I would check them and maybe rewind them.  Also check the traces on the little RX board for burn through.  Please do not ask me how I learned this.  Too embarrassing.  

                                  I agree that you should check the output of the LO.  There are two test points on your board where you can pick up the signal near the CMOS jumper.  You ought to be able to put a test wire on there and jumper it over to your antenna if you don't have a counter.  You should get all your signals as you change the switches.  

                                  Preston Douglas WJ2V

                                  -----Original Message-----
                                  From: g8voip <g8voi.reeves59@...>
                                  To: softrock40@yahoogroups.com
                                  Sent: Wed, Feb 3, 2010 5:58 pm
                                  Subject: [softrock40] Re: Si570 desoldering tips

                                   
                                  Hi Barry,

                                  An obvious question, but you are actually changing over and using the different RX filter and TX PA / filter modules when you are trying to receive on the higher frequency bands?

                                  To have good image rejection at the lower frequency suggests a problem with the phase splitter transformer on the RX filter module for the higher band.

                                  If you are listening for the LO on another receiver, remember the LO is running at 4 x the tuned frequency.

                                  To me it looks like you might have a combination of problems.

                                  You never did comment quite how you managed to destroy your v6.3 TX/RX in the first place. Knowing that might provide some clues to what has gone wrong, assuming that the board was completely working on all bands before the problem occurred.

                                  73, Bob G8VOI

                                  --- In softrock40@yahoogro ups.com, Barry Michels <bmichels@.. .> wrote:
                                  >
                                  > g8voip wrote:
                                  > > Hi Barry,
                                  > >
                                  > > Looking at your Rocky screen grab, you have no image rejection at all. Although that would not explain your apparent tuning problem above 10MHz, there is something pretty fundamental wrong with your receiver or PC / soundcard set up / configuration.
                                  > >
                                  > > I think you need to go back to basics and determine what is actually still working on your PCB.
                                  > >
                                  > > A good place to start is to find out exactly what the Si570 LO is doing.
                                  > >
                                  > > Do you have a frequency counter available you could hook up on say U9 pin 3 (clock into the 74HC74. If you use Fred PE0FKO's 'CFGSR' program, you can use the 'Tune' tab and select 1MHz steps and easily tune from 1MHz up to 30MHz, and you should be able to measure the Si570 output which will be at 4x the 'VFO' frequency ( 4 - 120MHz). That will confirm whether or not your LO is functioning correctly.
                                  > >
                                  > > If that is ok, then you will need to go on to the divider stage and confirm the two outputs to the RX mixer are correct. Access to an oscilloscope would be useful to see what is happening.
                                  > >
                                  > > In effect, follow the original build / checkout procedure.
                                  > >
                                  > > 73, Bob G8VOI
                                  > >
                                  > >
                                  > When I get some more time, I will run thru the checkout procedures.
                                  >
                                  > Below 10Mhz, I get very flat response (no hump in the middle, just a
                                  > thin spike) with great image rejection. The screenshots I posted only
                                  > look like that above 10Mhz (more like 10.2Mhz). Last time I checked, I
                                  > could hear the LO output on a nearby receiver even above 10Mhz. Will
                                  > try that again later.
                                  >
                                  > The only oscilloscope I have access to is my 32Mhz digital scope that
                                  > isn't worth much other than working with digital signals. It's an old
                                  > PCS32 parallel port model from Velleman.
                                  >
                                  >
                                  > Thanks for the help!
                                  > Barry - KE4JUC
                                  >

                                • Mr Doug -
                                  It might be a good idea to never connect theses boards to a high current supply and if you do make sure you have a fuse in the circuit. 1/4 A fast blow or
                                  Message 16 of 20 , Feb 3, 2010
                                  • 0 Attachment
                                    It might be a good idea to never connect theses boards to a high current supply and if you do make sure you have a fuse in the circuit.  1/4 A fast blow or less for the RX boards.

                                    If you put enough current through the toroid antenna coil primary to burn it open I am sure you would either see smoke or smell something. Even though it is relatively small wire it still would require a good deal of current to burn it open. I doubt a wall-wart would have enough current to do this. The antenna input is an isolated loop though and even if you had one side grounded the current applied would only go through that loop or to ground. 

                                    This is also a good reason to use back-to-back diodes across, and a lamp or fuse in series, with the antenna input.   


                                    From: Preston Douglas <pdouglas12@...>
                                    To: softrock40@yahoogroups.com
                                    Sent: Wed, February 3, 2010 6:23:08 PM
                                    Subject: Re: [softrock40] Re: Si570 desoldering tips

                                     

                                    Y'know, Barry, something you mentioned has been running around in my head (along with the squirrels!).  If the way you killed your 6.3 was by misconnecting the PA daughterboard so you shorted 12V to the amplifier board, you can't burn out chips that way.  They are isolated from the antenna inputs.  All you can do is burn out the secondary of the output transformer or the output inductor in the little RX filter.  I would check them and maybe rewind them.  Also check the traces on the little RX board for burn through.  Please do not ask me how I learned this.  Too embarrassing.  


                                    I agree that you should check the output of the LO.  There are two test points on your board where you can pick up the signal near the CMOS jumper.  You ought to be able to put a test wire on there and jumper it over to your antenna if you don't have a counter.  You should get all your signals as you change the switches.  

                                    Preston Douglas WJ2V

                                    -----Original Message-----
                                    From: g8voip <g8voi.reeves59@ tiscali.co. uk>
                                    To: softrock40@yahoogro ups.com
                                    Sent: Wed, Feb 3, 2010 5:58 pm
                                    Subject: [softrock40] Re: Si570 desoldering tips

                                     
                                    Hi Barry,

                                    An obvious question, but you are actually changing over and using the different RX filter and TX PA / filter modules when you are trying to receive on the higher frequency bands?

                                    To have good image rejection at the lower frequency suggests a problem with the phase splitter transformer on the RX filter module for the higher band.

                                    If you are listening for the LO on another receiver, remember the LO is running at 4 x the tuned frequency.

                                    To me it looks like you might have a combination of problems.

                                    You never did comment quite how you managed to destroy your v6.3 TX/RX in the first place. Knowing that might provide some clues to what has gone wrong, assuming that the board was completely working on all bands before the problem occurred.

                                    73, Bob G8VOI

                                    --- In softrock40@yahoogro ups.com, Barry Michels <bmichels@.. .> wrote:
                                    >
                                    > g8voip wrote:
                                    > > Hi Barry,
                                    > >
                                    > > Looking at your Rocky screen grab, you have no image rejection at all. Although that would not explain your apparent tuning problem above 10MHz, there is something pretty fundamental wrong with your receiver or PC / soundcard set up / configuration.
                                    > >
                                    > > I think you need to go back to basics and determine what is actually still working on your PCB.
                                    > >
                                    > > A good place to start is to find out exactly what the Si570 LO is doing.
                                    > >
                                    > > Do you have a frequency counter available you could hook up on say U9 pin 3 (clock into the 74HC74. If you use Fred PE0FKO's 'CFGSR' program, you can use the 'Tune' tab and select 1MHz steps and easily tune from 1MHz up to 30MHz, and you should be able to measure the Si570 output which will be at 4x the 'VFO' frequency ( 4 - 120MHz). That will confirm whether or not your LO is functioning correctly.
                                    > >
                                    > > If that is ok, then you will need to go on to the divider stage and confirm the two outputs to the RX mixer are correct. Access to an oscilloscope would be useful to see what is happening.
                                    > >
                                    > > In effect, follow the original build / checkout procedure.
                                    > >
                                    > > 73, Bob G8VOI
                                    > >
                                    > >
                                    > When I get some more time, I will run thru the checkout procedures.
                                    >
                                    > Below 10Mhz, I get very flat response (no hump in the middle, just a
                                    > thin spike) with great image rejection. The screenshots I posted only
                                    > look like that above 10Mhz (more like 10.2Mhz). Last time I checked, I
                                    > could hear the LO output on a nearby receiver even above 10Mhz. Will
                                    > try that again later.
                                    >
                                    > The only oscilloscope I have access to is my 32Mhz digital scope that
                                    > isn't worth much other than working with digital signals. It's an old
                                    > PCS32 parallel port model from Velleman.
                                    >
                                    >
                                    > Thanks for the help!
                                    > Barry - KE4JUC
                                    >


                                  • Barry Michels
                                    So, it wasn t the Si570... Turns out it was my sound card! Stupid me, I never thought about what damage the original event would have done to the sound card
                                    Message 17 of 20 , Feb 8, 2010
                                    • 0 Attachment
                                      So, it wasn't the Si570... Turns out it was my sound card! Stupid me,
                                      I never thought about what damage the original event would have done to
                                      the sound card that was connected.

                                      When I first built the RXTX, I could leave the 80M/40M filter in place
                                      and tune 160-10 in search of something interesting. I could see signals
                                      no matter where I went. If I found something, I 'd swap out the filter
                                      and get a stronger signal. Now, there's noise everywhere when the
                                      antenna isn't connected. With the antenna hooked up and the 80M/40M
                                      filter in place, everything under 30M shows up fine. But, even with the
                                      30M/20M/17M filter in place, all I get is noise above about 10.1Mhz.
                                      Last night, I was messing with my old TS-520 and using the RXTX as a
                                      tuning tool (TX low power on the TS-520 and tune the carrier until it
                                      showed up in the right spot in Rocky). With the antenna disconnected on
                                      the RXTX and noise all over the waterfall, as soon as I transmitted with
                                      the TS-520, all the noise went away and I could see my signal clearly.
                                      it got me thinking about the sound card. Maybe, now that it is damaged,
                                      it needs a minimum signal level to overcome the noise.

                                      I've connected it to my laptop and all bands are working. This old
                                      laptop can't process 96kbps, but 48 is ok. I'm WSPRing on 30M now (RX
                                      only). Guess it's time to buy a new sound card and this time install
                                      some isolation to protect it...


                                      Barry - KE4JUC
                                    • Peter
                                      I just noticed this post, so I apologize for being a couple of weeks late. I have a technique that works very well for unsoldering surface-mount parts. You
                                      Message 18 of 20 , Feb 15, 2010
                                      • 0 Attachment
                                        I just noticed this post, so I apologize for being a couple of weeks late. I have a technique that works very well for unsoldering surface-mount parts. You will need a hot air tool, I use a Weller Portasol butane soldering tool with a hot-air blower tip. You will also need an electric griddle, which you can get for about $20 at any discount store. You may want an infrared thermometer from Harbor Freight Tools or similar.

                                        Use the griddle as a hot plate to preheat the PCB. You'll have to watch the temperature to avoid damaging through-hole parts, this is where the thermometer comes in. When it reaches the desired preheat temperature, apply hot air. The solder holding the part you want to remove should melt almost instantly, and you can lift it right off. If you're replacing it, just place the new one while the PCB is still hot. More hot air to reflow the solder.

                                        Make sure you use a good hot air tool, NOT one of the little butane torches from harbor freight or similar, and use the hot air tip, NOT the blowtorch tip. You should see no flame.

                                        You can use an electric skillet instead of a griddle, but the griddle is easier to use because it doesn't have the sides.

                                        73,
                                        Pete NI9N
                                        www.lazydogengineering.com
                                        www.garage-shoppe.com

                                        --- In softrock40@yahoogroups.com, "barrymichels" <bmichels@...> wrote:
                                        >
                                        > I just received a shipment notification for the 2 Si570 samples I ordered 2 weeks ago. They should be here Wednesday. The Si570 that is in my RXTX won't let me tune over 10Mhz. What's the best method for removing the damaged chip without damaging the board?
                                        >
                                        > Thanks,
                                        > Barry - KE4JUC
                                        >
                                      • Francis DUPONT
                                        Hello Pete and All, ... late. ... parts. ... tool ... infrared ... You are right, It is a good method. But in the case of the Si570, I had proposed a
                                        Message 19 of 20 , Feb 15, 2010
                                        • 0 Attachment
                                          Hello Pete and All,

                                          On 2/15/2010 Pete NI9N wrote:

                                          >I just noticed this post, so I apologize for being a couple of weeks
                                          late.
                                          >I have a technique that works very well for unsoldering surface-mount
                                          parts.
                                          >You will need a hot air tool, I use a Weller Portasol butane soldering
                                          tool
                                          >with a hot-air blower tip. You will also need an electric griddle, which
                                          >you can get for about $20 at any discount store. You may want an
                                          infrared
                                          >thermometer from Harbor Freight Tools or similar.

                                          >Use the griddle as a hot plate to preheat the PCB. You'll have to watch
                                          >the temperature to avoid damaging through-hole parts, this is where the
                                          >thermometer comes in. When it reaches the desired preheat temperature,
                                          >apply hot air. The solder holding the part you want to remove should
                                          >melt almost instantly, and you can lift it right off. If you're
                                          >replacing it, just place the new one while the PCB is still hot.
                                          >More hot air to reflow the solder.

                                          >Make sure you use a good hot air tool, NOT one of the little butane
                                          >torches from harbor freight or similar, and use the hot air tip, NOT
                                          >the blowtorch tip. You should see no flame.

                                          >You can use an electric skillet instead of a griddle, but the griddle
                                          >is easier to use because it doesn't have the sides.

                                          You are right, It is a good method. But in the case of the Si570,
                                          I had proposed a MacGyver's style solution:

                                          On June 5 2008, in message # 21978, I wrote:


                                          > It happens that I have soldered mine this week. But before soldering,

                                          > I have devised a tip to unsolder it, just in case!

                                          > So here is the story:

                                          > The goal is to solder the Si570 at a small height (0.5mm for ex)
                                          above the board.

                                          > 1 - Put a small drop of solder on each of the 8 pads of the Si570.

                                          > 2 - Make a U shape with a piece of .05 mm wire. I took 2 inches of

                                          > wire wrapping wire (30 AWG/0.5 insulation diameter).

                                          > 3 - Put the U on the board, place and solder the Si570.

                                          > 4 - Remove the wire.

                                          Here use a monospaced font(fixed-width) like Courrier New etc.

                                          > ------X------
                                          > | |
                                          > X X
                                          > ____| |_______________X <- wire soldered on a
                                          > / | | pad or PCB track
                                          > | X Si570 X to avoid too
                                          > \____| |_______________X <- much move
                                          > | |
                                          > X X
                                          > | |
                                          > ------X------


                                          > "If" I have to remove the chip, I will place a piece of magnet wire

                                          > (30 AWG for ex) under it and unsolder one pad at a time by moving the

                                          > wire while the solder is melted. A well known solution.

                                          > (Once more it will be a good idea to fix one end of the magnet wire

                                          > by soldering it on the board.)


                                          As it costs nothing, I can't say it is my just my $.02 ... 73


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