[time-nuts] Low Phase Noise Amplifiers
I wonder if adding active bias feedback around the RF transistor to reduce the low frequency current variations would help. This is the classic PNP bias scheme which can be applied to BJT's or FET's. I have used it to successfully improve the phase noise on oscillators. Details from T.T. Ha, or Gonzales books on Amplifiers. Chuck, AF8Z ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Low Phase Noise Amplifiers
Am 11.01.20 um 15:36 schrieb Charles Clark: I wonder if adding active bias feedback around the RF transistor to reduce the low frequency current variations would help. This is the classic PNP bias scheme which can be applied to BJT's or FET's. I have used it to successfully improve the phase noise on oscillators. Details from T.T. Ha, or Gonzales books on Amplifiers. ... and available as cheap SOT343 chip from Infineon: BCR400W < https://www.infineon.com/dgdl/Infineon-BCR400W-DS-v01_01-en.pdf?fileId=db3a30431400ef68011407e93d8601a1 > I vaguely remember that its use in an oscillator has been patented. =8-() cheers, Gerhard ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Low Phase Noise Amplifiers
A VERY long time ago, it was discovered that simply degenerating a transistor with an emitter resistor makes a worthwhile improvement in 1/f noise. I want to say this was published in 1970 by Dick Baugh of HP but don't hold me to it. Note that the resistor was NOT bypassed: it's purpose was RF feedback, and any stabilization of bias current was incidental. The resistor value was a few dozens of ohms. That is not enough to do anything special in terms of stabilizing collector current. In oscillators, a designer might want to use a high performance bias stabilization scheme to minimize frequency drift (as opposed to noise). Various publications out of NIST (Fred Walls, et al) recommend using a transistor with high Ft vs the operating frequency to get low 1/f noise. This becomes more important when working at 100 MHz vs 10 MHz. As far as bias is concerned, the main emphasis seems to be on using a bias scheme that doesn't ADD noise to the amplifier. Rick N6RK On 1/11/2020 6:36 AM, Charles Clark wrote: I wonder if adding active bias feedback around the RF transistor to reduce the low frequency current variations would help. This is the classic PNP bias scheme which can be applied to BJT's or FET's. I have used it to successfully improve the phase noise on oscillators. Details from T.T. Ha, or Gonzales books on Amplifiers. Chuck, AF8Z ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] Low Phase Noise Amplifiers
FWIW, at the Arecibo Observatory all our cryogenic LNAs had bias stabilized with active stabilizers based on opamps. Since the opamps do not work at ~15K, bias connections to the drain and gate of the RF FETs were brought out separately from the RF connections, and the opamp circuitry was at room ambient temp. This approach would nicely stabilize both drain DC voltage and drain current over the whole temperature range from room ambient to 15K, which was handy for testing and monitoring LNA behavior during cool-down, which takes several hours (or more). Dana On Sat, Jan 11, 2020 at 9:40 AM Richard (Rick) Karlquist < rich...@karlquist.com> wrote: > A VERY long time ago, it was discovered that simply > degenerating a transistor with an emitter resistor > makes a worthwhile improvement in 1/f noise. I > want to say this was published in 1970 by Dick Baugh > of HP but don't hold me to it. Note that the resistor > was NOT bypassed: it's purpose was RF feedback, and > any stabilization of bias current was incidental. > The resistor value was a few dozens of ohms. That > is not enough to do anything special in terms of > stabilizing collector current. > > In oscillators, a designer might want to use a high > performance bias stabilization scheme to minimize > frequency drift (as opposed to noise). > > Various publications out of NIST (Fred Walls, et al) > recommend using a transistor with high Ft vs the > operating frequency to get low 1/f noise. This becomes > more important when working at 100 MHz vs 10 MHz. > As far as bias is concerned, the main emphasis seems > to be on using a bias scheme that doesn't ADD noise > to the amplifier. > > Rick N6RK > > On 1/11/2020 6:36 AM, Charles Clark wrote: > > I wonder if adding active bias feedback around the RF transistor to > > reduce the low frequency current variations would help. This is the > > classic PNP bias scheme which can be applied to BJT's or FET's. I have > > used it to successfully improve the phase noise on oscillators. Details > > from T.T. Ha, or Gonzales books on Amplifiers. > > > > Chuck, AF8Z > > > > > > ___ > > time-nuts mailing list -- time-nuts@lists.febo.com > > To unsubscribe, go to > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > > and follow the instructions there. > > > > ___ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. > ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
[time-nuts] The difficulty of low noise measurements
Fellow time-nuts, As I now have a hydrogen maser sitting here, it triggers me to measure things. Essentially I try to measure the stability of the maser, and well, that will be very hard since one need very quiet reference sources to do that, but that then triggers the question of how quiet are my sources anyway. So, I decided to start a measuring campaign to figure this out and to measure both ADEV stability and phase-noise stability, and with that build reference for the future of what tools to use for a particular measurement. It also has the side-effect that it forces me to get more of my lab in order. Anyway, so as I measure, I find that I have problem getting good ADEV measure. I operate my TimePod such that I can do the three-corner hat. I see some disturbance in the ADEV with a re-occurrence of about 0.5 s and that part of the ADEV never converges to something useful even when doing a 48 h test. At the upper end, the BVAs I use for reference is clearly drift limiting, so I have to restort to Hadamard to see somewhat beyond the drift limit, but that is not helpful. So, I decide to setup phase-noise measurments and do that while the BVAs settles, as drift does not obstruct the phase-noise as much as ADEV. As I measure phase-noise of my 5065, OSA8600, OSA8601 and EFOS10 it becomes very clear that the EFOS10 does not have even the same shape of the phase noise as the other three, and being far worse than the reference BVAs despite the output oscillator is a OSA8600 and for higher frequencies this should completely dominate the phase-noise behaviors. It comes to the point where I decide I need to break a measurement and investigate this deeper. So, a tell-take of the phase-noise plot is that it has some strange spikes at about 0.1 Hz and 0.3 Hz. It then has ripples that eventually goes smooth as the filter-bandwidth of the phase-noise measurement increase, so it just becomes an almost smooth noise energy which slopes down. To put this in context, it is about 30 dB higher than expected when it is as worst. Surely, this is something other than the intended noise of the maser. It now becomes time to shift view again, remember we started with long-term stability in ADEV and HDEV and then shifted to phase-noise. So, I shift over to view the phase data, and in particular the residue phase as the linear phase-slope has been removed (press 'r' in the phase view of TimeLab). I then see regular phase-spikes at 10 s appart. I measure these to be in the range of 10 ps high. For a while I have my suspicions falsly on the PFD detector in the PLL, as I have a compassionate dislike/hate for them, but Bob and John talked me out of doing that hack right now. Better first collect more evidence. Then it hits me. Could it be... what happens in 10 s interval here, the only process going on with that type of rate is the logging of data from the maser. So, I pull the cable and start new measurements... and that is it. I now have ADEV operating smoothly in the region where it was clearly fighting, and I have a nice smooth 1/tau slope down into the abyss from the maser, clipping the 1E-13 line at tau=3 s at which time it is so quiet that as it progresses into the E-14 range of measures it becomes hard to measure using BVAs. Phase-noise measurement is a spitting image of the BVA I have, and those matches up with TvBs measurements of OSA8600. Now, I have already took the precaution to sprinkle the cables with common mode suppression chokes. I have not used any fancy power-supplies, but just a standard HP/Agilent bench supply for the BVAs, and they are being fed from the same supply even. The EFOS-B masers was shipped with optical isolators for RS232, and this may help for common mode and "ground loop" but not for the transitions themselves. I will have to investigate this further to figure out how the main action that makes the bursts of data on the serial port to appear in the measured 5 MHz, and then how to isolate it. Until then I now know I need to pull that cable for precision measurements, but that is not a satisfactory solution as I want that log-data as an intact series. I send this note, to also indicate the importance of monitoring the data in different views, that phase-noise measurement may be giving a significant clue as to what goes on, and then the phase (and/or frequency) view to further locate it. Amplitude noise and pure spectrum analysis may have been additional methods to also consider, just to illustrate that one needs a wide selection of tools to locate the root cause of a disturbance and mitigate it. I've heard people say "I'm only interested in stability", well, that may be the end goal of the measurement, but to get there and to improve the setup other measurement discipilnes may be necessary. Thanks goes to Bob Camp and John Miles as discussion partners, and to Ole Petter Rønningen who helped me a lot with the EFOS10 maser. I will continue my measurement campaign, and see if there is more thin
Re: [time-nuts] The difficulty of low noise measurements
At that level of the dungeon, you have a choice. You can either bolt *everything*, including the chair you sit on and the pencil you write with, together with copperstraps to get a common potential. Or you can arrange *all* your cables and other metalic connections (cabinets touching because the insulating legs were lost years ago?) to form a Directed Acyclic Graph[1]. Many people overlook that if you run 2 m cable to an opto isolator you still have a 2m antenna. It may no longer bother you in terms of EMC, but it can still bother you in terms of EMI[2]. I have on my ever-growing TODO list to test if serial-BLE adapters are any good. Has anybody tried that yet ? But foremost: Use individual float-charged lead-acid for power-supplies. so that you can unplug from the grid when you *really* want to measure. For trivial loads, like the BVA, motorcycle batteries from "BilTema" are fine, but for larger capacity batteries you get what you pay for, so contact somebody who knows what they are selling and will not send you batteries which has been sitting on some shelf for 3 years[3]. Poul-Henning [1] That's a fancy way of saying "no loops anywhere". [2] Note that the Maxim RS232 switched-capacitor level conveters are noisy as hell, as are many USB-RS232 adapters. [3] Or check if you can inherit at ton of OPzS from friends in telecom, but pay attention to the hydrogen. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] The difficulty of low noise measurements
On 1/11/20 9:57 AM, Poul-Henning Kamp wrote: At that level of the dungeon, you have a choice. You can either bolt *everything*, including the chair you sit on and the pencil you write with, together with copperstraps to get a common potential. Or you can arrange *all* your cables and other metalic connections (cabinets touching because the insulating legs were lost years ago?) to form a Directed Acyclic Graph[1]. Two theoretically isolated 2m tall, 1m deep racks separated by 1cm is a 2nF capacitor. At 10 MHz, that's a not so huge series 9 ohms in your DAG. > I have on my ever-growing TODO list to test if serial-BLE adapters > are any good. Has anybody tried that yet ? That is an interesting idea. I wonder if you could make a short distance fiber connection by just having your notional 10MHz source directly drive the diode, with maybe a series resistor (so you don't need a power supply). You could almost certainly do this for an RS232 output. I wonder if there's a "no power required" approach for the receiving end. or a low noise RS232-fiber adapter that steals power for the Rx side by using the Tx signals. No level shifters and charge pumps. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] The difficulty of low noise measurements
Hi, On 2020-01-11 18:57, Poul-Henning Kamp wrote: > At that level of the dungeon, you have a choice. > > You can either bolt *everything*, including the chair you sit on > and the pencil you write with, together with copperstraps to get a > common potential. Mesh Bonding Network [1] > > Or you can arrange *all* your cables and other metalic connections > (cabinets touching because the insulating legs were lost years ago?) > to form a Directed Acyclic Graph[1]. Isolated Bonding Network [1] > > Many people overlook that if you run 2 m cable to an opto isolator > you still have a 2m antenna. It may no longer bother you in terms > of EMC, but it can still bother you in terms of EMI[2]. Indeed. Then, being a ham, antennas is familiar to me. [2] > > I have on my ever-growing TODO list to test if serial-BLE adapters > are any good. Has anybody tried that yet ? My recommendation is to build on the ESP32, then you can get serial port to WiFi directly and the entry level price is small enough and there exist a community of things. It is competent enough that you may let it do some of the processing for you. > But foremost: Use individual float-charged lead-acid for power-supplies. > so that you can unplug from the grid when you *really* want to > measure. The EFOS10 maser has UPS and batteries, sure they are not prime but I can for shorter measurements just go and unplugg power if needed. > > For trivial loads, like the BVA, motorcycle batteries from "BilTema" Indeed. I will setup a separate 24 V charger chain for the BVAs and a few other things, as I want UPS for them too. > are fine, but for larger capacity batteries you get what you pay > for, so contact somebody who knows what they are selling and will > not send you batteries which has been sitting on some shelf for 3 > years[3]. I have a ham friend that is fairly well into batteries as he needs to for his telecom installations. > > Poul-Henning > > [1] That's a fancy way of saying "no loops anywhere". Indeed. I know that, but not all is familiar with the computer science term. > > [2] Note that the Maxim RS232 switched-capacitor level conveters > are noisy as hell, as are many USB-RS232 adapters. I guess that is why I had two common mode chokes on the serial cable, one at the USB adapter end and the other at the maser. > > [3] Or check if you can inherit at ton of OPzS from friends in > telecom, but pay attention to the hydrogen. Good idea. It happens that we get these. [1] ITU-T Rec. K.27 https://www.itu.int/rec/T-REC-K.27/en [2] My club SK0UX http://sk0ux.se Cheers, Magnus ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] The difficulty of low noise measurements
In message <454f3441-6479-cc54-7c1c-55f61b5fa...@earthlink.net>, jimlux writes: >That is an interesting idea. I wonder if you could make a short distance >fiber connection by just having your notional 10MHz source directly >drive the diode, with maybe a series resistor (so you don't need a power >supply). You could almost certainly do this for an RS232 output. I >wonder if there's a "no power required" approach for the receiving end. There are people abusing TOSLINK hardware along those lines. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] The difficulty of low noise measurements
In message <643136da-9599-18ca-be85-ffa62ab04...@rubidium.se>, Magnus Danielson writes: >> I have on my ever-growing TODO list to test if serial-BLE adapters >> are any good. Has anybody tried that yet ? >My recommendation is to build on the ESP32, then you can get serial port >to WiFi directly and the entry level price is small enough and there >exist a community of things. It is competent enough that you may let it >do some of the processing for you. The reason I noted "BLE" and not "WIFI" is the lower power: Less RF can never be a bad thing in these circumstances. >The EFOS10 maser has UPS and batteries, sure they are not prime but I >can for shorter measurements just go and unplugg power if needed. Have you tried it ? How is the electrical grid configured on your rocks with respect to PEN ? Do you have a ground electrode at each house or does the utility provide the PEN potential via their cable ? In the former case: How old are the one at your house ? If more than few decades old, it may not provide good contact any more. In the latter case, your PEN will be very noisy, and if you think you can hammer a 2.5m hard copper "electrode" into some kind of wet underground, a galvanic trafo on the mains might be relevant. >I have a ham friend that is fairly well into batteries as he needs to >for his telecom installations. LVDC people are much better at this than UPS people, the latter often think that batteries only need to last "until the diesel kicks in" and that is a very different application. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
[time-nuts] OCXO Advice
I’d like some advice about the OCXO’s I’m using. They’re UCT 108663-10, double oven, purchased from the auction site. The attached graph shows the frequencies of 2 very similar circuits compared to the 10 MHz output of an LPRO-101. At 7.6 hours, the blue line design switches from GPS PLL to hold mode. The red line design is still in GPS PLL (not equipped with hold mode) and it’s not fair to compare the blue line noise to the red line noise after 7.6 hours. Prior to 7.6 hours, comparison shows the blue line design has more noise. Both OCXO’s have 100’s of hours of use and were on for about 50 hours prior to what shows in the graph. I don’t know if the range of slow noise I see from both OCXO’s is typical. Any other choices for OCXO that would be quieter and affordable? I measured the frequency control voltage at the OCXO pins and found it stable within 100 uV. Scale factor of the OCXO is about 0.9 Hz per V or 0.9 mHz per mV and the drift would require more control voltage drift than what I see. I also checked the 12 V supply to the OCXO and found it stable within 100 uV. Both OCXO’s are similarly mounted. Looks like ambient temperature is a factor. ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
Re: [time-nuts] OCXO Advice
Hi Bob, On 2020-01-11 23:42, Bob Quenelle wrote: > I’d like some advice about the OCXO’s I’m using. They’re UCT 108663-10, > double oven, purchased from the auction site. The attached graph shows the > frequencies of 2 very similar circuits compared to the 10 MHz output of an > LPRO-101. At 7.6 hours, the blue line design switches from GPS PLL to hold > mode. The red line design is still in GPS PLL (not equipped with hold mode) > and it’s not fair to compare the blue line noise to the red line noise after > 7.6 hours. Prior to 7.6 hours, comparison shows the blue line design has more > noise. Both OCXO’s have 100’s of hours of use and were on for about 50 hours > prior to what shows in the graph. I don’t know if the range of slow noise I > see from both OCXO’s is typical. Any other choices for OCXO that would be > quieter and affordable? > > I measured the frequency control voltage at the OCXO pins and found it stable > within 100 uV. Scale factor of the OCXO is about 0.9 Hz per V or 0.9 mHz per > mV and the drift would require more control voltage drift than what I see. I > also checked the 12 V supply to the OCXO and found it stable within 100 uV. > Both OCXO’s are similarly mounted. Looks like ambient temperature is a factor. I would recommend you to load the data into TimeLab, as it is a good tool to analyze data, and more importantly it could allow you to see the plots progress as data comes in. Somewhere here I should have some OSA8663 for which the UCT108663-10 is probably made to be second source variant off. It doesn't seem to be terriby bad oscillators, but it may help to create ADEV plots and compare to other oscillators to see how they compare. Cheers, Magnus ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
[time-nuts] 5065A phase-noise
Fellow time-nuts, As I have been measuring this, I've been surprised by the behavior differences of my 5065A rubidiums. One of them has a modernization so the 00105-6013 oscillator is replaced by a 10811 oscillator with associated support-board. This retrofit is known as 05065-6097 [1]. Now, the phase-noise plots show two main things: For the 00105-6013 assembly, the far out noise-floor is around -154 to -155 dBc. This fine. However, at around 1 kHz is the 1/f flicker noise corner, providing considerable noise. I have measured both with closed and open-loop, but the noise is the same, so this comes from the oscillator. The loop opens just at the EFC input of the oscillator. The transition to 1/f^3 corner comes further down, but not really bad location, the flicker phase sticks out. So, then for the replacement kit, the 10811 is really good, now isn't it? Well... turns out that the wide noise is around 12-13 dB worse than the old 00105-6013, it's only as one gets to the closed in noise that it raises up, and it is quite respectable. I've not measured this open-loop yet. So, none of them is stellar in the phase-noise department. I've noticed that the 0565-6097 has a rather crude adaptation of the sine into the TTL input of a 7474 flip-flop setup to act as divide by 2. You have a series inductor and capacitor into a voltage divider-chain that sets the midpoint. I am not quite sure that this provides good noise properties as it squares up. I've not spotted likely reasons for problems yet in the 00105 oscillator, most transistors seems to have pretty OK emitter degeneration which should help to keed flicker noise down. Have a look at ponder over these things. Cheers, Magnus [1] http://www.ko4bb.com/manuals/HP_5065A_A10_quartz_oscillator_replacement_kit_1981_restored.pdf ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
[time-nuts] Any (relatively) cheap TSIP GPS receivers without WNRO out there?
I have an NTP server that uses a Trimble GPS receiver (Accutime 2000, p/n - 39091-00, ROM 3.06) that has fallen prey to WNRO. Just wondering if there are any Trimble (or other brand) receivers that have TSIP serial comms that are available for a reasonable price? Any help appreciated. Skip Withrow ___ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
