Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
- Loopantennas
- Messages
Search
|
Re: Steve ugly-build LZ1AQ using MPS2222 sweeps uploaded to Photos
On Thu, Jul 18, 2024 at 05:51 PM, <biastee@...> wrote:
I have done a lot of 2-Tone IM measurement of lo-Z Input loop amplifiers, LZ1AQ, Wellbrook, my own designs etc. and have not encountered a degradation of the IP2 and IP3 because of 2-in-series inverse parallel silicon small signal diodes. At a small loop in shortcut mode, connected to a lo-Z input transimpedance amplifier, the voltages across the amplifier input are too low to get into the nonlinear area of the diodes V-I curve. So i do not yet see a compelling reason to replace protection diodes for 1 ct/piece by PIN Limiter Diodes for 2 $ per piece, unless somebody proves me wrong by measurement or can show me a significant improvement of the OIP2 and OIP3. ?
Owen duffy's silicon diode criticism refers to an Hi-Z E-field antenna, a MiniWhip. In this case you have to reckon with higher voltages and electrostatic transient voltages to GND. With this type of antennas, unbiased 1N4148 or BAV99 diodes are indeed not recommended. At e-field antennas like whips and dipoles i normally use biased silicon diodes or ultra low capacitance ESD Diodes for signal line and antenna protection, as for example:
?
?
?
?
>I haven't encountered these diodes being used in commercial equipment's limiters.
?
I can direct you to many professional and military grade active antenna designs from Rohde & Schwarz (e.G. HE-series) and from other renownded manufacturers that use biased silicon diodes for input protection of the? amplifier. By the way, i also count Wellbrooks ALA1530 series antenna as commercial equipment.
?
regards
Fred |
|
Re: Steve ugly-build LZ1AQ using MPS2222 sweeps uploaded to Photos
Martin,
?
When you have been experimenting with pairs of tuned loops at close spacing the degree of magnetic coupling K between them rouglhly? 0.75 or higher which means that
two resonant circuits form a common resonance frequency.
If we change the strength of magnetic coupling we can change the frequency of the common resonance frequency.
In adition to that common lower frequency f1 a second higher resonance frequency f2 is formed as well. This is clearly visible in simulation and also in the impedance measurement. At frequency f2 the currents in both loops flow in push-pull.
The lower frequency: f1= 1/2x pi x sqrt ((La+Lm) x C), where Lm=kxLa - mutual inductance?
Let's assume K=1, then Lm=L? and? f1= 1/2 pi sqrt 2Lx C?
For uncoupled loop: f0= 1/2 pi sqrt La x C, then? f1= f0/sqrt 2 showing how much the frequency can be lowered vs. uncoupled loop with resonant frequency f0.
Also a second effect comes into play - additional capacitive coupling ( mutual capacity ) between loops.
Thus, the common frequency: f1= 1/2x pi x sqrt ((La+Lm)x(C-Cm)), where Cm - mutual capacitance
If we bring two tuned loops closer together Lm and Cm increase equally. So, in practice f1 will change lesser than calculated 1/sqrt2.
The behaviour of two tuned loops corresponds largerly to behaviour of dual-circuit band filter.
To observe critical coupling in MLA, the two tuned loops would have to be several meters apart, starting to see two humps in the frequency characteristics.
?
Raphael |
|
Re: Steve ugly-build LZ1AQ using MPS2222 sweeps uploaded to Photos
> But the expected voltages across a small loop in lo-Z shortcut?
> mode are much lower than at fullsize or hi-Z e-field antennas. > But in principle i absolutely agree with your advice to avoid? > antiparallel diodes. Better use biased diodes or bidirectional? > ultra low capacitance TVS Diodes. From Martin's comment that two 82pF "improved the performance on the upper HF bands", I infer that the Wellbrook also has an impedance that rises with frequency. ?
/g/loopantennas/message/19758 ?
What are the RF performances of biased (forward?) diodes and bidirectional ultra low capacitance TVS Diodes in limiter applications? I haven't encountered these diodes being used in commercial equipment's limiters.
A PIN diode limiter HSMP-4820 turns on at 7 dBm and has IP3 = 15 dBm.?
?
> However, this is not the case with broadband untuned loops,?
> and they seem to be relatively insensitive to their surroundings. ?
Aren't the untuned loops also inadvertently tuned? Eg. the LZ1AQ loop is parallel resonated by two 100 pF, and the Wellbrook by the two 82 pF caps. Maybe it the amp's low input Z that renders these inadvertently tuned loops insensitive?
|
|
Re: Steve ugly-build LZ1AQ using MPS2222 sweeps uploaded to Photos
Thanks for that.
?
One thing that struck me about the loop spacing, is that when using a tuned loop, the Q is much higher and there will be grater interaction with nearby objects, such as a test excitation loop.?
?
I have noticed this when experimenting with pairs of tuned loops, at close spacing the interaction is quite significant. However, this is not the case with broadband untuned loops, and they seem to be relatively insensitive to their surroundings.
?
Regards,
?
Martin
?
?
On Thu, Jul 18, 2024 at 02:57 PM, Raphael Wasserman wrote:
Please find the requested info. in my album titled "IEEE Standard, 1931 Radio Engineers' Handbook".? |
|
Re: Added photo album IEEE Standard, 1931
#photo-notice
The first page states that it was the "Standards on Radio Receivers", Institute of Radio Engineers, New York, 1938. IIRC Harold Wheeler was the chairman of that group. 73? John? KC0G / M0KCY
On Thursday, July 18, 2024 at 09:45:17 AM CDT, Group Notification <[email protected]> wrote:
Raphael Wasserman <wassermanr46@...> added the photo album IEEE Standard, 1931 : Radio Engineers' Handbook _._,_.? |
|
Photo Notifications
#photo-notice
Group Notification
The following photos have been uploaded to the IEEE Standard, 1931 photo album of the [email protected] group. By: Raphael Wasserman <wassermanr46@...> Raphael Wasserman <wassermanr46@...> updated the photo album IEEE Standard, 1931 : Radio Engineers' Handbook |
|
Added photo album IEEE Standard, 1931
#photo-notice
Group Notification
Raphael Wasserman <wassermanr46@...> added the photo album IEEE Standard, 1931 : Radio Engineers' Handbook |
|
Added Folder /IEEE Standard, 1931
#file-notice
Group Notification
Raphael Wasserman <wassermanr46@...> added folder /IEEE Standard, 1931 Description: |
|
Re: Wellbrook traced-out loop amp schematics uploaded to Photos
Hi Steve,
?
There has been a bit of variation in the build of the Wellbrooks.
?
The one I took apart had two 82pF capacitors, each one connected across half of the transformer primary. I suspect they were originally intended to provide better rejection of strong FM broadcast signals, but it was a "happy accident" that they also improved the performance on the upper HF bands.
?
The core material was closer to N30 (u' 4,300) (Epcos ) rather than 73 mix (u' 2,500) , which would tie in with the FG100LN additional transformer, that used something like the higher permeability x38 material (u' 10,000). My thinking is that the Epcos N30 cores would also have possibly been easier to obtain in the UK, when the Wellbrook was initially designed, as RS were pretty much the only big electronic parts supplier at the time, and was used by nearly everyone.
?
Using back to back diodes across the loop terminals is fine, however that is not true for other types of antenna, with higher feed point impedances.
?
I have tried CPH3910 fets and they work, but a lot depends upon your requirements for frequency coverage.
?
Regards,
?
Martin |
|
Re: Wellbrook traced-out loop amp schematics uploaded to Photos
I wish those diagrams works, if my Wellbrook ALA100M someday stop working I will be totally lost, the urban noise in my location is very high. ? It would be nice that somebody in this group could sell them mounted or al least in kit , if allowed. |
|
Re: Wellbrook traced-out loop amp schematics uploaded to Photos
I don't know much about semiconductor physics. So, my layman's guess is that the 2SK715 must be larger / wider than common JFETs because Yfs scales with gate width. A 90s era JFET intended for AM tuner, BF862A has yfs = 45 mA typ.
Not JFET, but useful for comparison. The more recent ePHEMTs have even higher yfs. The ATF-55143 0.5 x 400 um (Lgate x Wgate) is 220 mS typ., while the ATF-54143 at 0.5 x 800 um is 410 mS typ.
?
73,
Chin-Leong Lim, 9W2LC
?
p/s where can I find your measurement data on the mini-whip? thanks. |
|
Re: Steve ugly-build LZ1AQ using MPS2222 sweeps uploaded to Photos
On Wed, Jul 17, 2024 at 11:53 PM, <biastee@...> wrote:
Additionally, the Wellbrook teardown showed two pairs of anti-parallel 1N4148 at the input which would have made Owen Duffy sneered in disgust :-). Read Duffy's scathing dismissal of "diodes connected to antennas": ? Thats a valid point for most antennas. But the expected voltages across a small loop in lo-Z shortcut mode are much lower than at fullsize or hi-Z e-field antennas. But in principle i absolutely agree with your advice to avoid antiparallel diodes. Better use biased diodes or bidirectional ultra low capacitance TVS Diodes.
?
regards
Fred |
|
Re: Steve ugly-build LZ1AQ using MPS2222 sweeps uploaded to Photos
> ?downloaded IEC 315-1 standard. Item 20 - "Antenna Simulation Networks?
> Artficial Aerials )" basically covers three type of antennas: open-wire? > antenna 100 KHz - 30 MHz, rod and telescopic antennas? ?
The above comment made me realized the Indian version is far less descriptive than my photocopy of the (cutthroat) original. IMHO, the missing information renders the Indian version useless. So, please view the following:
?
> I found in IEEE Standard, 1961 the details about the?
> recommended dummy antenna impedance that mimics? > open-wire antenna: Za=1500 Ohms @ 500 KHz, 200? > Ohms @ 2 MHz and 400 Ohms @ 10 MHz. ?
Can you kindly point me to the document? Thanks.?
?
> I thought Steve Ratzlaff did some "unpotting" amongst others
> Maybe we need to find out what they found. > I remember seeing results from at least two "unpottings" > I would think an optimized impedance matching over a relativity?
> large frequency range was super critical. ?
Martin's website (now defunct?) pointed to .?According to the teardown, there is no LPF in the Wellbrook amp, save for a pair of 82 pF caps at the input. Hence, there's NO impedance matching at the input that tracks the loop impedance changing with frequency. AFAIK, only LZ1AQ with its LPF has an input impedance that track the loop's - albeit a "happy accident" in Martin's word.?
Additionally, the Wellbrook teardown showed two pairs of anti-parallel 1N4148 at the input which would have made Owen Duffy sneered in disgust :-). Read Duffy's scathing dismissal of "diodes connected to antennas": ?
Duffy's concerns over PN diodes (e.g. 1N4148) causing intermodulation can be addressed by substituting with long lifetime PIN diodes. So, as (ex) application engineer for PIN diode limiters for over 2 decades, I must lend my wholehearted disapproval, too! :-)?
?
> Ikin knew all that decades ago !!!!!!!!!!
?
I do NOT relish in building and evaluating so many designs - LZ1AQ, PA0FRI, M1GEO, & MLA-30, etc. So, if someone can furnish me with quantitative data showing that Ikin's Wellbrook is the undisputed performance leader, then I will throw the other designs out of the window and fully embrace the Wellbrook! :-)
Digression: Almost all active whips have high input Z. Interestingly, there is contrarian design which sports low input Z using feedback in order to achieve high IP3 (53 dBm). Hmmm... low Z like LZ1AQ minus the LPF :-) https://analog-electronics.tudelft.nl/downloads/NordholtActiveAntenna.pdf |
|
Re: Wellbrook traced-out loop amp schematics uploaded to Photos
The 2SK715 jfet used in the original has a typical yfs of 50 mS--which is much higher than any other jfet--somehow Andrew Ikin discovered them. When the circuit was first traced out I built a copy and verified it worked the same as an original Wellbrook ALA1530LN. I experimented with other jfets, none came close to working like the 2SK715. I also experimented with different paralleled jfets--I tried one, two, three in parallel (4 in parallel each side on the original). The input impedance steadily decreases as jfets are added--Andrew Ikin apparently decided 4 in parallel gave him the desired input impedance with the input transformer turns ratio he used, for the small one-meter diameter loop. 2SK715s were still available when I first traced the circuit out, though they were already listed as "obsolete", and I was fortunate to get some for my tests.
toggle quoted message
Show quoted text
73, Steve AA7U On 7/17/2024 5:44 PM, Steve Ratzlaff via groups.io wrote:
|
|
Wellbrook traced-out loop amp schematics uploaded to Photos
开云体育I made a folder in Photos just now and uploaded the schematics
for ALA1530LN, ALA100LN, FLG100LN that I've traced out in the
past. There should be complete info there for anyone to be able to
make their own. The original leaded 2SK715 jfets are now
unobtainable, but there are two SMD equivalents that work just as
well, with the 2SK932 preferred since its typical yfs is the same
as the 2SK715. One should be able to make one of these for around
$50--maybe less. On eBay I see a Chinese seller claims to be
selling an ALA1530LN or an ALA100LN for a very high price of
around $250, for the loop amp alone! I have no idea if it's a
direct clone of the Wellbrook. /g/loopantennas/album?id=296460.??? The name of the folder is "Wellbrook official traced-out schematics ALA1530LN, ALA100LN, FLG100LN". 73, Steve AA7U On 7/17/2024 2:15 PM, Paul V Birke via
groups.io wrote:
|
|
Re: 2e0ero Magnetic Loop Antenna
i was just grumbling that i could not add a picture .. all good
toggle quoted message
Show quoted text
soooo ... anybody who wants to take a look .. look my qrz page dg9bfc sigi Am 14.07.2024 um 15:53 schrieb Simon: Hi Sigi |
to navigate to use esc to dismiss