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

Re: Dual Transistors for Active Negative Inductors

Expand Messages
  • David
    Digikey and Mouser also have dual transistors, in both surface mount and leaded format, in prices that vary between more expensive and less expensive than the
    Message 1 of 19 , Oct 13, 2012
      Digikey and Mouser also have dual transistors, in both surface mount and leaded format, in prices that vary between more expensive and less expensive than the auction site. I'm not recommending one source over the other and I haven't compared specs with the part Chris found. I just did a quick search to see what options I had and was surprised at how much discrete stuff is still available from current production.

      Dave Sarraf


      >><christrask@> wrote:

      While searching for dual transistors that are suitable for use with the active negative inductor circuits I came up with, I came across some interesting offerings on eBay. Specifically, there is a seller in Canada who has MBT3904DW1 and MBT3906DW1 devices available at very reasonable prices. And there was also a somewhat interesting FMW2 device, which is a pair of 2CS2412 NPN transistors with a common base connection. These are not monolithic duals, but are instead a pair of individual die on a single carrier. Usually, devices such as this are comprised of two die that are immediately adjacent on the wafer, so their characteristics are fairly well matched, certainly better than using two separate devices.

      >>
    • Judi Black_Lake
      The MBT3904 series is available from ON Semiconductor and you can order samples. ORDER:http://www.onsemi.com/PowerSolutions/product.do?id=MBT3904DW1
      Message 2 of 19 , Oct 14, 2012
        The MBT3904 series is available from ON Semiconductor and you can order samples.


        ORDER:
        http://www.onsemi.com/PowerSolutions/product.do?id=MBT3904DW1

        DATASHEET:
      • Chris Trask
        ... I looked through Digikey and Mouser and came up with the BFM505 and DMC206E2 as being suitable candidates for the active negative inductor circuits. Not
        Message 3 of 19 , Oct 14, 2012
          >
          > Digikey and Mouser also have dual transistors, in both surface mount
          > and leaded format, in prices that vary between more expensive and less
          > expensive than the auction site. I'm not recommending one source over
          > the other and I haven't compared specs with the part Chris found. I
          > just did a quick search to see what options I had and was surprised at
          > how much discrete stuff is still available from current production.
          >

          I looked through Digikey and Mouser and came up with the BFM505 and DMC206E2 as being suitable candidates for the active negative inductor circuits. Not exactly cheap, but they both look like good performers.

          I have difficulty recommending any of the NEC/CEL monolithic duals as I've noticed over the years that their offerings are not exactly stable.

          Chris
        • claudio
          Hi Chris, I ve been out of the topic because of my job, but now I m coming back again. Are you looking for a NPN/PNP matched pair ? Or simply two matched NPN
          Message 4 of 19 , Oct 15, 2012
            Hi Chris,
            I've been out of the topic because of my job, but now I'm coming back again.
            Are you looking for a NPN/PNP matched pair ? Or simply two matched NPN for the Linvill's type NIC ?
            During the gaps of my job I found very interesting SiGe transistors from Infineon (we are using them at 15 Ghz) namely the BFP740 or BFP420, they have high Ft and low noise, I suspect that they will work well in some of the schemes we've worked up to now including the Linvill's one. I suspect this because in my simulations I can see the input noise rising according to the gain fall due to Ft limits...will keep you informed.
            Anyway I'm in stall because of noise and IMD...!

            73s Claudio
            --- In loopantennas@yahoogroups.com, Chris Trask <christrask@...> wrote:
            >
            > >
            > > Digikey and Mouser also have dual transistors, in both surface mount
            > > and leaded format, in prices that vary between more expensive and less
            > > expensive than the auction site. I'm not recommending one source over
            > > the other and I haven't compared specs with the part Chris found. I
            > > just did a quick search to see what options I had and was surprised at
            > > how much discrete stuff is still available from current production.
            > >
            >
            > I looked through Digikey and Mouser and came up with the BFM505 and DMC206E2 as being suitable candidates for the active negative inductor circuits. Not exactly cheap, but they both look like good performers.
            >
            > I have difficulty recommending any of the NEC/CEL monolithic duals as I've noticed over the years that their offerings are not exactly stable.
            >
            > Chris
            >
          • Roberto
            Ciao ! I would try to build a breadboard of small antenna @ 27 MHz and I would like to undertake two different options for active matching, to be compared as
            Message 5 of 19 , Oct 16, 2012
              Ciao !
              I would try to build a breadboard of "small antenna" @ 27 MHz and I would like to undertake two different options for active matching, to be compared as follows :

              Option a) using MAX2180 (AM section) antenna amplifier

              Option b) using BFM505 for practical NIC circuits shown by Chris:
              Fig.5 on pag.5 of "Active Negative Inductor" Report
              http://home.earthlink.net/~christrask/Active%20Negative%20Inductors.pdf
              Fig.6 on pag.5 of "Linvill feedback amplifier" Report
              http://home.earthlink.net/~christrask/Linvill%20Feedback%20Amplifier.pdf

              I would be truly glad of receiving your advise

              Warmest ciao !
              Roberto


              --- In loopantennas@yahoogroups.com, "claudio" <aetna50@...> wrote:
              >
              >
              >
              > Hi Chris,
              > I've been out of the topic because of my job, but now I'm coming back again.
              > Are you looking for a NPN/PNP matched pair ? Or simply two matched NPN for the Linvill's type NIC ?
              > During the gaps of my job I found very interesting SiGe transistors from Infineon (we are using them at 15 Ghz) namely the BFP740 or BFP420, they have high Ft and low noise, I suspect that they will work well in some of the schemes we've worked up to now including the Linvill's one. I suspect this because in my simulations I can see the input noise rising according to the gain fall due to Ft limits...will keep you informed.
              > Anyway I'm in stall because of noise and IMD...!
              >
              > 73s Claudio
              > --- In loopantennas@yahoogroups.com, Chris Trask <christrask@> wrote:
              > >
              > > >
              > > > Digikey and Mouser also have dual transistors, in both surface mount
              > > > and leaded format, in prices that vary between more expensive and less
              > > > expensive than the auction site. I'm not recommending one source over
              > > > the other and I haven't compared specs with the part Chris found. I
              > > > just did a quick search to see what options I had and was surprised at
              > > > how much discrete stuff is still available from current production.
              > > >
              > >
              > > I looked through Digikey and Mouser and came up with the BFM505 and DMC206E2 as being suitable candidates for the active negative inductor circuits. Not exactly cheap, but they both look like good performers.
              > >
              > > I have difficulty recommending any of the NEC/CEL monolithic duals as I've noticed over the years that their offerings are not exactly stable.
              > >
              > > Chris
              > >
              >
            • Chris Trask
              ... would ... compared as ... I have done a considerable amount of modeling to determine where the active negative inductors I described would work best. The
              Message 6 of 19 , Oct 16, 2012
                >
                > Ciao !
                > I would try to build a breadboard of "small antenna" @ 27 MHz and I
                would
                > like to undertake two different options for active matching, to be
                compared as
                > follows :
                >
                > Option a) using MAX2180 (AM section) antenna amplifier
                >
                > Option b) using BFM505 for practical NIC circuits shown by Chris:
                > Fig.5 on pag.5 of "Active Negative Inductor" Report
                > http://home.earthlink.net/~christrask/Active%20Negative%20Inductors.pdf
                > Fig.6 on pag.5 of "Linvill feedback amplifier" Report
                > http://home.earthlink.net/~christrask/Linvill%20Feedback%20Amplifier.pdf
                >
                > I would be truly glad of receiving your advise
                >

                I have done a considerable amount of modeling to determine where the
                active negative inductors I described would work best. The usable bandwidth
                of the circuit of Fig. 5 is directly proportional to the embedding
                impedance. In other words, the usable bandwidth increases as the embedding
                impedance increases. There is a limit to this imposed by the transistors,
                and I found the for HF frequencies an embedding impedance of about 5 ohms
                works best.

                The combination I would recommend would be to use the negative inductor
                circuit in a balanced configuration, coupled in place of the varactors in
                Fig. 4 of my earlier paper on varactor-tuned loop matching networks:

                http://www.home.earthlink.net/~christrask/Paper011.html

                To make a balanced negative inductor for this application, add a third
                winding to transformer T2 of Fig. 5, using a trifilar twist of wires to
                ensure good coupling and balance. Be careful to get the correct phase
                connections when doing this. It might be prudent for me to add this
                application to the negative inductor paper as that was the intended use from
                the beginning.

                Chris
              • Chris
                ... I just uploaded a schematic into the group files section to help illustrate this, and the announcement appeared about a minute ago. I ll add this
                Message 7 of 19 , Oct 16, 2012
                  <<SNIP>>

                  >
                  > To make a balanced negative inductor for this application, add a third
                  > winding to transformer T2 of Fig. 5, using a trifilar twist of wires to
                  > ensure good coupling and balance. Be careful to get the correct phase
                  > connections when doing this. It might be prudent for me to add this
                  > application to the negative inductor paper as that was the intended use from
                  > the beginning.
                  >

                  I just uploaded a schematic into the group files section to help illustrate this, and the announcement appeared about a minute ago. I'll add this schematic to the active negative inductor paper later, along with further comments about the selection of transistors and other details.

                  Chris
                • Roberto GIUSTO
                  I have just created a LTSpice simulation using two 2N2222 transistorsLet me know your comments and suggestions Warmest ciao !                      
                  Message 8 of 19 , Oct 25, 2012
                  I have just created a LTSpice simulation using two 2N2222 transistors
                  Let me know your comments and suggestions

                  Warmest ciao !

                                            Rob
                • Roberto GIUSTO
                  Dear Chris, the series-resistor of 100 Ohm gives Collector currents of 60 mA, therefore, leaving the supply voltage at 7 V and increasing the series-resistor
                  Message 9 of 19 , Oct 25, 2012
                  Dear Chris,
                  the series-resistor of 100 Ohm gives Collector currents of 60 mA, therefore, leaving the supply voltage at 7 V and increasing the series-resistor on the collector up to 470 Ohm, we get 13 mA  (see the enclosed simulation results of DRAFT2)

                  Regarding the Source resistance of 5 Ohm, I have replaced the former small-signal Current source (Itest1) with a small-signal Voltage source V2 in series with the required 5 Ohm resistance. The result is here enclosed in DRAFT3.

                  It seems this circuit shows "ideal inductive" behaviour at lower frequencies under 10 MHz, ...and "non-Foster" behaviour in the band 30MHz-50MHz..isn't it ?

                  Warmest regards,
                                      Roberto

                  --- Ven 26/10/12, Chris Trask <christrask@...> ha scritto:

                  Da: Chris Trask <christrask@...>
                  Oggetto: Re: [loopantennas] Re: Dual Transistors for Active Negative Inductors
                  A: loopantennas@yahoogroups.com
                  Data: Venerdì 26 ottobre 2012, 00:25

                   

                  >
                  > I have just created a LTSpice simulation using two 2N2222 transistorsLet
                  me
                  > know your comments and suggestions
                  >

                  What bias voltage and current are you using?

                  Looks like the solid line is signal current and the dashed line is
                  phase.

                  If you place a 5-ohm resistor in series with the signal source you
                  should see better results. My simulations showed this impedance
                  to be close to ideal.

                  Chris

                • Roberto
                  Dear Chris, I ve built the circuit using two 2N3904, it s oscillating! Thus, back to my LTSpice simulation now for checking transient behaviour (that is in
                  Message 10 of 19 , Nov 25 1:56 AM
                    Dear Chris,
                    I've built the circuit using two 2N3904, it's oscillating!

                    Thus, back to my LTSpice simulation now for checking transient behaviour (that is in time_domain for large signal) I found that instability affects such circuit configuration.

                    Therefore, if you simulate the behaviour in the frequency domain you get just small-signal response of the circuit and you cannot see this instability.

                    Look at my simulation result NIC_2N3904_INSTABILITY.pdf in the "Files" section, folder : LTspice_simulates_2-NPN_Neg_INDUCTANCE

                    I will send this result also by email, for your convenience...

                    Warmest ciao !




                    --- In loopantennas@yahoogroups.com, "Chris Trask" <christrask@...> wrote:
                    >
                    > >
                    > > Ciao !
                    > > I would try to build a breadboard of "small antenna" @ 27 MHz and I
                    > would
                    > > like to undertake two different options for active matching, to be
                    > compared as
                    > > follows :
                    > >
                    > > Option a) using MAX2180 (AM section) antenna amplifier
                    > >
                    > > Option b) using BFM505 for practical NIC circuits shown by Chris:
                    > > Fig.5 on pag.5 of "Active Negative Inductor" Report
                    > > http://home.earthlink.net/~christrask/Active%20Negative%20Inductors.pdf
                    > > Fig.6 on pag.5 of "Linvill feedback amplifier" Report
                    > > http://home.earthlink.net/~christrask/Linvill%20Feedback%20Amplifier.pdf
                    > >
                    > > I would be truly glad of receiving your advise
                    > >
                    >
                    > I have done a considerable amount of modeling to determine where the
                    > active negative inductors I described would work best. The usable bandwidth
                    > of the circuit of Fig. 5 is directly proportional to the embedding
                    > impedance. In other words, the usable bandwidth increases as the embedding
                    > impedance increases. There is a limit to this imposed by the transistors,
                    > and I found the for HF frequencies an embedding impedance of about 5 ohms
                    > works best.
                    >
                    > The combination I would recommend would be to use the negative inductor
                    > circuit in a balanced configuration, coupled in place of the varactors in
                    > Fig. 4 of my earlier paper on varactor-tuned loop matching networks:
                    >
                    > http://www.home.earthlink.net/~christrask/Paper011.html
                    >
                    > To make a balanced negative inductor for this application, add a third
                    > winding to transformer T2 of Fig. 5, using a trifilar twist of wires to
                    > ensure good coupling and balance. Be careful to get the correct phase
                    > connections when doing this. It might be prudent for me to add this
                    > application to the negative inductor paper as that was the intended use from
                    > the beginning.
                    >
                    > Chris
                    >
                  • Chris Trask
                    ... This may be due to mismatch of the transistors. If the emitter resistance of the second device is larger than that of the first device you ll end up with
                    Message 11 of 19 , Nov 25 5:51 AM
                      >
                      >Dear Chris,
                      > I've built the circuit using two 2N3904, it's oscillating!
                      >
                      > Thus, back to my LTSpice simulation now for checking transient behaviour
                      > (that is in time_domain for large signal) I found that instability affects
                      > such circuit configuration.
                      >
                      > Therefore, if you simulate the behaviour in the frequency domain you get
                      > just small-signal response of the circuit and you cannot see this instability.
                      >
                      > Look at my simulation result NIC_2N3904_INSTABILITY.pdf in the "Files"
                      > section, folder : LTspice_simulates_2-NPN_Neg_INDUCTANCE
                      >
                      > I will send this result also by email, for your convenience...
                      >

                      This may be due to mismatch of the transistors. If the emitter resistance of the second device is larger than that of the first device you'll end up with a negative input resistance. Try adding a small resistance in series with the emitter of the first device.

                      Chris
                    • Roberto GIUSTO
                      Dear Chris,                 the circuit is oscillating even using its SPICE model (for transient simulation), actually in these simulations the two
                      Message 12 of 19 , Nov 25 9:03 AM
                        Dear Chris,
                                         the circuit is oscillating even using its SPICE model (for transient simulation), actually in these simulations the two transistors are identical.

                           Looking forward receiving your comments and suggestions
                              Warmest regards,
                                                           Roberto 

                        --- Dom 25/11/12, Chris Trask <christrask@...> ha scritto:

                        Da: Chris Trask <christrask@...>
                        Oggetto: Re: [loopantennas] Re: Dual Transistors for Active Negative Inductors
                        A: loopantennas@yahoogroups.com, loopantennas@yahoogroups.com
                        Data: Domenica 25 novembre 2012, 14:51

                         


                        >
                        >Dear Chris,
                        > I've built the circuit using two 2N3904, it's oscillating!
                        >
                        > Thus, back to my LTSpice simulation now for checking transient behaviour
                        > (that is in time_domain for large signal) I found that instability affects
                        > such circuit configuration.
                        >
                        > Therefore, if you simulate the behaviour in the frequency domain you get
                        > just small-signal response of the circuit and you cannot see this instability.
                        >
                        > Look at my simulation result NIC_2N3904_INSTABILITY.pdf in the "Files"
                        > section, folder : LTspice_simulates_2-NPN_Neg_INDUCTANCE
                        >
                        > I will send this result also by email, for your convenience...
                        >

                        This may be due to mismatch of the transistors. If the emitter resistance of the second device is larger than that of the first device you'll end up with a negative input resistance. Try adding a small resistance in series with the emitter of the first device.

                        Chris

                      • Chris Trask
                        ... What is the frequency of the oscillation? My simulation shows that there is a moderate potential for instability around 40MHz or so using the 2N3904. The
                        Message 13 of 19 , Nov 25 11:35 AM
                          >
                          >Dear Chris,                 
                          >the circuit is oscillating even using its SPICE model (for transient
                          >simulation), actually in these simulations the two transistors are
                          >identical.
                          >   Looking forward receiving your comments and suggestions     
                          >

                          What is the frequency of the oscillation? My simulation shows that there is a moderate potential for instability around 40MHz or so using the 2N3904. The solution to this is to place a resistor in parallel with the reference inductor.


                          Chris Trask
                          N7ZWY / WDX3HLB
                          Senior Member IEEE
                          http://www.home.earthlink.net/~christrask/
                        Your message has been successfully submitted and would be delivered to recipients shortly.