Charles P. Steinmetz wrote: > Bruce wrote: > >> A circuit schematic for a current feedback triple with reasonably low >> noise and distortion is attached. > > One caution regarding the 100 uH inductor (L3) -- many inductors of > this value exhibit self-resonance below 10 MHz, so some care may be > necessary in selection. > > Best regards, > > Charles > EPCOS produce a suitable inductor with its first SRF around 20Mhz or so. The inductor may be omitted if noise and loop gain isnt too critical. Bruce

cfo wrote: > On Mon, 26 Mar 2012 08:11:16 +1300, Bruce Griffiths wrote: > > >> Bruce Griffiths wrote: >> > >> A circuit schematic for a current feedback triple with reasonably low >> noise and distortion is attached. >> >> Bruce >> > Thanx Bruce > > I'm an analog noob , so i have some questions. > > If i multiply that schematic by 8 , could i then just add the Tbolt input > at all V1's ? > > The input capacitance may be a little high, however one could approximately compensate for it in a narrow band application. > Should R15 + C11 + C12 also be multiplied ? > > One RC decoupling circuit like this for each amplifier is advisable to reduce crosstalk. > Is V2 a 12v supply (i suppose so) ? > Yes the schematic was taken from an LTSpice simulation. > Could i replace the 2N3904 with BC850LT1 (smd) , and the 2N3906 with > BC856ALT1 (smd) , else i have some BC337/BC327 - TO92 (non smd) > > No, the input and output capacitances of those transistors are probably a little too high. > Would one chan work on a perf-board with wires (proof of concept) , or > would it be to "RF sensitive" ? > > Its usually better to construct it over a ground plane "rats nest" style. A piece of unetched PCB works well. > regards > CFO > > > Bruce

On 3/25/2012 9:54 PM, gary wrote: > MMBD914 !=1n914. > 1n914BWTm i.e. using a suffix, is something I haven't seen before, but > technically > 1N914BWT != 1n914. That is, in the strict sense, the 1n914 has to be a > diode in that glass package. As long as we're being pedantic, you're wrong. What you say is only true if it is a JEDEC 1n914 that you're talking about. "1n914" cannot be trademarked or copyrighted. A manufacturer is perfectly free to make a device in a non-glass package and call it a 1n914, which means it _is_ a 1n914, as long as they stay away from JEDEC. Then again, following your lead in being impractically, completely, worthlessly pedantic, it's a JEDEC 1N914, not a 1n914. In the strict sense, the latter cannot exist under JEDEC. "No one cares" is probably an understatement.

I was only kidding a bit, since for a lot of the most common JEDEC and also non-JEDEC conventional (as opposed to SMD) discrete components they are similar parts with a similar numbering in SMD, and the 1N914 is one of them - so I found it not the best example :) . However, the same argument could be applied to the MMBT3904... in the same sense, != 2N3904 since in the strict sense it is not TO-92. Anyway I was not intending to generate any deep discussion about this matter :) Regards, Javier El 26/03/2012 03:54, gary escribió: > MMBD914 !=1n914. > 1n914BWTm i.e. using a suffix, is something I haven't seen before, > but technically > 1N914BWT != 1n914. That is, in the strict sense, the 1n914 has to be a > diode in that glass package. > > > On 3/25/2012 5:48 PM, Javier Herrero wrote: >> El 26/03/2012 02:35, gary escribió: >>> I forgot to mention that those old jedec part numbers specify a >>> package and electrical limit under one part number. That is, you can't >>> find say a 1n914 in SMD, but you can find direct equivalents with >>> other numbers. You will find supply houses listing SMD versions of >>> jedec parts, but technically that is not correct. >> Oh, you can :) MMBD914 (SOT-23), 1N914BWT /SOD-523F) and some more. >> >> Regards, >> >> Javier >> >> _______________________________________________ >> time-nuts mailing list -- time-nuts@febo.com >> To unsubscribe, go to >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > >

It would be interesting if the FTS-1050A curves were also plotted. Joe From: John Ackermann N8UR <jra@febo.com> To: Discussion of precise time and frequency measurement <time-nuts@febo.com> Date: 03/25/2012 03:54 PM Subject: Re: [time-nuts] Distribution amp - Use a video amp unit ? Sent by: time-nuts-bounces@febo.com I haven't measured the video amps, but here are plots of an HP 5087A, TADD-1, and Spectracom 8140 tap unit for a baseline: http://febo.com/pages/amplifier_phase_noise/ John ---- Tom Knox said the following on 03/25/2012 03:48 PM: > > Has anyone measured Phase Noise on any of these distribution amps? By looking at how widely the PN specs vary on the application specific Symmetricom and HP DA's it would be interesting to see how well these video da's perform in comparison. > Thomas Knox > > > >> Date: Sun, 25 Mar 2012 12:28:56 -0700 >> From: pete@petelancashire.com >> To: time-nuts@febo.com >> Subject: Re: [time-nuts] Distribution amp - Use a video amp unit ? >> >> Sounds like Paul and I have similar DA's or will soon when a Leitch gets to the >> front porch next week. Its bandwidth is spec'd at 30 MHz. >> >> The only question I have on using DA's is the effect of them being designed to >> work in a 75 ohm environment. >> >> One thing to watch out for is what cards they have in them. >> >> -pete >> >> >> >> On Sun, Mar 25, 2012 at 11:39 AM, paul swed<paulswedb@gmail.com> wrote: >>> Well the easy answer. >>> The older 1984 video amps would roll off above 6 MC or so. >>> However as the IC technology took over they easily could do 30 Mhz. >>> I have used Grass valley 8500 series for at least 10-15 years and the work >>> very well. >>> I also have sets of Leitch DAs that are fine. >>> Picked up the whole tray for some silly amount of $ 20 I think. >>> I prefer the simple 8601 DAs but have 8604 delay days and they can all be >>> made into simple DAs with a few jumpers that are on the boards. >>> I have also used a rgb da 6 output for each channel. No issue at all. >>> One comment you would want to insure the amps are ac coupled and clamping >>> turned off since there is no sync signal to establish the clamp level. >>> Regards >>> Paul >>> WB8TSL >>> >>> On Sun, Mar 25, 2012 at 2:14 PM, Bruce Griffiths<bruce.griffiths@xtra.co.nz >>>> wrote: >>> >>>> cfo wrote: >>>> >>>>> I need a distribution amp for my Tbolt , with 6..8+ outputs >>>>> >>>>> The TAPR-1 is NA , and the successor status is ???. >>>>> >>>>> I need it for distributing 10Mhz ref-clock to Counters >>>>> (HP,Racal,Phillips), a Rigol Sig-gen& maybe Oscilloscopes etc. >>>>> >>>>> >>>>> It's for "better than ocxo/hobby use" , and not Tnut (ps accuracy). >>>>> >>>>> I have seen suggestions for using a video distributiution unit. >>>>> But in my part of the world (EU) , there isn't a lot on "eby". >>>>> I have seen some RGB amps , and was wondering if i could use those ? >>>>> >>>>> Could i use a 1:6 RGB unit for : Tbolt on R , FEI-5680A on G - and ?? on >>>>> B ? >>>>> >>>>> I'm not experienced (yet) in the analog domain. >>>>> >>>>> If i go (simple diy) Would this one work (also wo. the trafo's) ? >>>>> http://www.oz9fw.dk/PDF%**20files/rubidium.pdf< http://www.oz9fw.dk/PDF%20files/rubidium.pdf> >>>>> >>>>> I do have some trafo's on an old 10Mhz 8-chan Hub , that i might be able >>>>> to salvage if needed. >>>>> >>>>> Thanx for any input/hints >>>>> >>>>> CFo - T-Nut beginner Denmark. >>>>> >>>>> >>>>> >>>>> ______________________________**_________________ >>>>> time-nuts mailing list -- time-nuts@febo.com >>>>> To unsubscribe, go to https://www.febo.com/cgi-bin/** >>>>> mailman/listinfo/time-nuts< https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts> >>>>> and follow the instructions there. >>>>> >>>>> >>>>> >>>> Winding your own transformers isnt particularly difficult if you have >>>> suitable (binocular) cores. >>>> The discrete amplifier shown has a class AB output stage, lower distortion >>>> is possible if a class A output stage is used (requires an extra resistor >>>> between the output stage emitters and an additional diode in series with >>>> the LED). >>>> With a Class A output stage the 39 ohm resistors should be replaced by a a >>>> pair of 100 ohm resistors to match the load. >>>> Significantly lower distortion is possible if feedback from the output to >>>> the input stage is used. >>>> If the feedback gain is unity (or less) at dc higher AC gains won't >>>> significantly degrade the close in phase noise. >>>> A different amplifier topology using the same number of transistors will >>>> allow a higher reverse isolation to be achieved. >>>> >>>> One problem with video distribution units (aside from the relatively high >>>> phase noise floor) is the limited output drive available. For some purposes >>>> 16-20dBm outputs are desirable. >>>> >>>> Bruce >>>> >>>> >>>> ______________________________**_________________ >>>> time-nuts mailing list -- time-nuts@febo.com >>>> To unsubscribe, go to https://www.febo.com/cgi-bin/** >>>> mailman/listinfo/time-nuts< https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts> >>>> and follow the instructions there. >>>> >>> _______________________________________________ >>> time-nuts mailing list -- time-nuts@febo.com >>> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>> and follow the instructions there. >> >> _______________________________________________ >> time-nuts mailing list -- time-nuts@febo.com >> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.