The AM Forum
March 28, 2024, 09:25:43 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: RF choke core material  (Read 25045 times)
0 Members and 1 Guest are viewing this topic.
PA4WM
Member

Offline Offline

Posts: 62


« on: December 15, 2012, 06:16:18 AM »

Our transmitters are full of 'em... 2.5mH rf chokes.
I found some retailers on the internet, most of them in the US.
Apparantly there are only a few manufacturers that make these.
Bourns/Miller and Hammond were the names that I found, and maybe they are even the same.

http://www.tubesandmore.com/products/P-C1535B

In the specs I noticed that the core material is made of ferrite.
I have some old rf chokes that are wound on a ceramic core.

Does the ferrite core of those 2.5mH chokes have any disadvantage in our applications?
I know ferrite has a wanted influence on the inductance, but ferrite and RF are not always close friends... Grin

I plan to order some, but am a bit worried about the core.
Should I?


Martin, PA4WM
Logged

PA4WM / WM2J
W4AMV
Member

Offline Offline

Posts: 672


« Reply #1 on: December 15, 2012, 10:06:54 AM »

Hi Martin,

At high current the core can saturate and the permeability can drop and so does the inductance. So that is one issue. The old National and other chokes using scramble winding like the plate chokes at 2.5 mH avoid the use of ferrite. Their self resonant frequency is high and out of band where it does exist. The ferrite loaded core, depending on its permeability could be troubling. However, very broadband RF ferrite loaded inductors are provided with operation to 40 GHz! So, the short answer Martin is YES, exercise care in the selection of ferrite material. After building a ferrite loaded inductor, measure its properties and that will assist you in determining its applications. Material properties of the various ferrite types are on the WEB. Read and understand how those properties apply to your application is helpful.

Alan
Logged
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #2 on: December 15, 2012, 06:38:31 PM »

Chokes for what purpose? Nobody uses a 2.5mH choke for anything in current production except an antenna "safety" choke or to raise a GG amp grid above ground. Those are all pi wound phenolic or ceramic core and vary from 1 to 2.5mH.

Plate chokes use ceramic or hard Teflon (Teflon and fiberglass). There have been a few practical uses of ferrite loaded plate chokes but it seems to be a lost or understudied art; I certainly blew up several attempting it. When Ameritron took an old QST design and after several attempts got it to include WARC it was more economical to just buy theirs.

The problem with pi wound chokes is resonances in or near hambands, some which werent available when the old ones were designed. They still work fine in an untuned low level stage in the 160-40M range. A much lower value can be used in a tuned stage to 10M but should be tested for resonances and/or swamped with ~ a 3K carbon resistor.

The National R-175A was probably the best of the old plate chokes as it covered the traditional 160-10M bands while the older R-175 blew up on 15. The wire is only #30 so its not for QRO.

The best deal on the planet for an amp plate choke is the big Ameritron which is still only $20 which is also sold by RF Parts at $5 more.
It will handle well over 3KW CW/SSB 160-10M including WARC. Ive pushed a 1000W AM carrier plus sidebands thru one during some testing on a customers amp and it survived nicely on all bands as well as 5KW PEP on SSB.

Carl
Logged
PA4WM
Member

Offline Offline

Posts: 62


« Reply #3 on: December 16, 2012, 04:10:47 AM »

Chokes for what purpose?
Carl

I should have been more clear here Wink
I didn't mean the plate chokes. Those wimpy 2.5mH pi chokes from Hammond or Bourns are to small for that use anyway.

Was thinking about low power driver stages. 6AG6 - 6146 (or 807).
I would like to use a choke between the g1 bias resistors and ground, in order to meassure the negative voltage.
Another choke has to come in the plate leads of these driver tubes. If it were for a single band setup, I would use a parallel  resonance circuit.
But I have a multiband driver in mind. The plate of the first stage will be untuned, and only the final will be tuned to resonance, with the plate shunt fed.

So for these low voltage and current stages I need to have a few of those small 2.5mH chokes.
Would the Hammond choke mentioned, with the ferrite core, be suitable for this...?

Martin
Logged

PA4WM / WM2J
N8ETQ
Member

Offline Offline

Posts: 795


Mort


« Reply #4 on: December 16, 2012, 07:04:23 AM »


[/quote]



So for these low voltage and current stages I need to have a few of those small 2.5mH chokes.
Would the Hammond choke mentioned, with the ferrite core, be suitable for this...?

Martin

[/quote]

Hey Martin,

     Yes,  I have an old Ceramic job that the leads broke off
hanging from the lamp over my bench just for this purpose.
Works fine.

/Dan


* choke.jpg (458.91 KB, 1824x1368 - viewed 2000 times.)
Logged
w7fox
Member

Offline Offline

Posts: 102


« Reply #5 on: December 16, 2012, 05:44:58 PM »

Plate choke comment.

I just finished replacing the plate choke on my homebrew pair of 1625's.  It was one of those 1 or 2.5 mh three pie types that came out of some old military surplus.  It was getting quite toasty on ten meter AM, so I would see a wisp of smoke now and then.  It just had to go, so I started poking around in the junk box and found some solenoid wound chokes that came out of some scrapped ART-13's.  Not the big one on the 813, but for something else, maybe a safety HV choke on the antenna circuit.  What a difference it made, power out on ten meters went from 45 watts to 65 watts, and no more smoke.  I guess I know where that 20 watts was going.  I did use a grid dip meter to check resonances, and I would recommend anyone else using a mystery choke do the same.  I had to remove a few turns to get the resonance away from 15 meters.  Good luck fellow builders, 73.

Chris
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #6 on: December 16, 2012, 06:04:48 PM »

Martin, I understood your question and in the end some measurements go a long way to figuring out what is required and what we have on hand. Not addressing plate chokes, I have some of the chokes that Carl mentioned. I'll measure a few and post the results. One as I recall is 1mH the other 2.7mH R-152 from National. I suspect at low frequency, under 1 MHz, they will measure as advertized and eventually become self resonant. At that point, we have a SHUNT capacitor with some Q, that is some series equivalent R. Its that R value that can lead to loss of power particularly at higher operating frequency, see prior post. So, what we believe are inductive chokes, at high operating frequency are actually capacitors with DC continuity! Once you have data on these old chokes, you may be able to construct same with ferrite types with the appropriate perm or AL to get the low frequency  inductance needed.       
Logged
N8ETQ
Member

Offline Offline

Posts: 795


Mort


« Reply #7 on: December 16, 2012, 06:08:03 PM »

Hey Chris,

   "Classic Tale" indeed, I melted, literally a 1" dia. Teflon
Rod with a hand wound Plate choke I made for an 88 Ish
mcy. FM TX years ago. Also HB. Essentially a class C doubler
from an RT 68! 4CX250's..  Used to inject a little 19kc
to turn on the "Stereo Led"..   Man I need therapy!

73

/Dan


Plate choke comment.

I just finished replacing the plate choke on my homebrew pair of 1625's.  It was one of those 1 or 2.5 mh three pie types that came out of some old military surplus.  It was getting quite toasty on ten meter AM, so I would see a wisp of smoke now and then.  It just had to go, so I started poking around in the junk box and found some solenoid wound chokes that came out of some scrapped ART-13's.  Not the big one on the 813, but for something else, maybe a safety HV choke on the antenna circuit.  What a difference it made, power out on ten meters went from 45 watts to 65 watts, and no more smoke.  I guess I know where that 20 watts was going.  I did use a grid dip meter to check resonances, and I would recommend anyone else using a mystery choke do the same.  I had to remove a few turns to get the resonance away from 15 meters.  Good luck fellow builders, 73.

Chris
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #8 on: December 16, 2012, 09:17:34 PM »

Here is some measured data on a "classic" 1 mH choke. A National R152 choke has similar characteristics but a lower resonant frequency, near 1.3-1.4 MHz. In any case you can see that above ~ 3.5 MHz, this 1 mH choke is NO longer inductive and represents a shunt Capacitance with a series resistance value that is shown in blue in the plots ~ 120 ohms at 30 MHz.  So if you build a ferrite equivalent with the appropriate material selection, you might use the values highlighted in these plots as a typical target.  Hope this provides some direction to your post.


* a_1mH_choke.JPG (64.88 KB, 960x720 - viewed 1768 times.)

* b_1mH_choke.JPG (56.92 KB, 960x720 - viewed 1513 times.)
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #9 on: December 16, 2012, 11:07:21 PM »

I made an error in recording the angle of the impedance at resonance. The corrected plot below. The remainder of the 1 mH Z RFC plot is ok. Again, above 4 MHz or so, this value of the RFC looks like a capacitor (of-course with DC continuity) and is reasonably small, under 2 pF. The C recorded is corrected by the stray C of the probe used in the measurement.


* c_1mH_choke.jpg (76.98 KB, 960x720 - viewed 1541 times.)
Logged
VE3LYX
Member

Offline Offline

Posts: 769


Crystals are from the stone age


WWW
« Reply #10 on: December 19, 2012, 12:57:18 PM »

I make my own, almost always. I used to use wooden dowel, 1/4 inch but after building the single 6293 Rf PA I found on long transmit at around 800 volts it would get warm to the point of giving a wee puff of smoke from the wooden centre. . It is in the plate as a plate choke. I remade it on an old 5000 watt wire wound resistor after severing and removing the wire wound resistance part. It works good. I have tortured it to be sure. No smoke which is always good.
This morning I wound one for the osc plate supply in my homebrew T1154 simplified AM transmitter circuit.  I had a handful of ceramic instrument resistors that have been in my junk box for probably 20 plus years. I unwound one and rewound it with many feet of #28 copper. I put on enuf wire to be SURE it will not be resonant anywhere near where I am using it. I put on lots of wire then add another 30 to 50 turns. The value? I dont know. They work. I have probably 10 or twelve of them in service here. Both the "woodys" and a few of the ceramic ones. Wish I had lots of the old ones around but they are really hard to find and to buy. By the time you ship them, too many $$$$$$$$$$$. These come out of the junk box for free and work just fine.
Don VE3LYX
Logged

Don VE3LYX<br />Eng, DE & petite Francais
PA4WM
Member

Offline Offline

Posts: 62


« Reply #11 on: December 20, 2012, 03:54:37 PM »

Thanks for all the replies, and Alan for the plots.
I think I understand what you are saying.
I must admit, I never gave rf chokes much attention. But now, in the phase of aquiring parts, it forces me to think about them...
I will use those 2.5mH pie wound chokes (the ones that have a ferrite core and are still being sold) in the low current driver stages g1 and plate.
I re-planned the 6146b's plate circuit, and will use a series fed resonant tank instead of a choke and shunt fed tank.
Those 2.5mH chokes are selfresonant on 1.3MHz. I hope this is far enough away from 1.8MHz to cause troubles.


Martin
Logged

PA4WM / WM2J
W4AMV
Member

Offline Offline

Posts: 672


« Reply #12 on: December 21, 2012, 11:28:25 AM »

Martin, I just visited your first post web site. I looked at the RFC they are advertising... There is NO FERRITE used in that element. These units are wound on phenoloic or bakelite or variations of that material. The wire I suspect as in the unit I measured is Litz wire to maintain a low DC resistance and reaonable Q. Beyond that, the prior posts suggesting the construction of this type choke are all on target. Without the use of ferrite, the key will be achieving a reasonable inductance with a self resonant frequency that is reasonable. And note from the plots and as a preferred element, the distributive capacitance of the windings should be small... a few pF is key. Even if it is somewhat larger, the self capacitance of the RFC can be absorbed into the matching network or added to the device capacitance. Of course, it can't get to large! Finally on plate chokes, the handbook and remarks form other posts are key. On the plate choke (shunt feed case vs. series feed case)... The SERIES resonant frequency, not the anti-resonannt or parallel resonant frequency is important. I did NOT measure or report that value of series resonant freq. on the small choke I measured. The plate chokes which may be physically LARGE can exhibit a stray series C value which if series resonant "in-band" could cause high circulating currents in the choke. More to the point, since the plate choke network is NOT a large parallel resonant Z, its going to upset the PI or ELL matching network significantly! As the handbook explains, plate chokes should be first measured installed with the plate side of the choke shorted to the supply line side and a GDO used to search for resonance. I am sure there are experts here who can shed light on how to properly construct these type chokes.
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #13 on: December 21, 2012, 01:03:59 PM »

Here is a useful catalog found on the web.

http://www.peakbagging.com/Electronic/JamesMillen_1972.pdf

Has the SRF for a number of the classic chokes! As well, the DCR and current.

73' Alan
Logged
PA4WM
Member

Offline Offline

Posts: 62


« Reply #14 on: December 22, 2012, 03:15:47 AM »

That is a nice Millen catalog!

The posted link to the Hammond choke however says the core isd made of ferrite:

Impregnated with a moisture and fungus resistant varnish.
1-1/2 long axial leads.
Wound for low distributed capacity by using 3-Pi windings.
Inductance: 2.5 mH5% tolerance
Q (min): 106
Test Frequency (KHz): 250
Self Resonant Min. Frequency (MHz): 1.3
Maximum DC current resistance (Ohms): 9
Maximum DC current (mA): 160
Coil diameter (in): .469
Core Length (in): .875
Core Material: Ferrite

It's just the ferrite part that I have doubts about.
Thanks to your graphs and explanation Alan, I'm getting the idea of a choke even being capacitive on higher frequenties. And the more I dig into this, I wonder why all the vintage driver transmitters are full of those 2.5mH chokes. They could, and can cause a lot of trouble...
I'm currently rethinking my idea of a 3 (lower) band transmitter, to avoid rf chokes where possible.

Martin
Logged

PA4WM / WM2J
W4AMV
Member

Offline Offline

Posts: 672


« Reply #15 on: December 22, 2012, 08:20:34 AM »

The view of the picture (to my eyes) does not look like Ferrite. OK, a magnified view shows a tubular cardboard over a black (ferrite??) rod OR a carbon film/composition resistor?? So I really can't say for sure and if it is ferrite it would be nice if they specified the permeability of the material! In any case, as you see in the Millen catalog, all the classic chokes you reference in the original post are operated above their parallel self resonant frequency for most of the ham bands. That is OK... This is not an issue. As long as the self capacitance of the choke is not dominant in the circuit application. Also in the catalog, note the materials specified are not ferrite. So I suspect the author of the advertisement is in error. My opinion. Again, if the choke self capacitance is small, the reactance at the operating frequency is high and effect of the choke is in play, that is: PASS DC DO NOT DISTURB (block) RF.

Plate chokes and the possibility of SERIES resonance is another matter and their is good cite that I found that treats that subject. The KK5DR has notes on plate chokes that are ferrite loaded, however, there are comments that imply issues! So, I believe in the end careful measurements and slow experiments to qualify a high power choke are in order. All of mine have been wound on ceramic, delron, and in one case Teflon (not a wise choice).
Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #16 on: December 22, 2012, 11:10:34 AM »



http://www.peakbagging.com/Electronic/JamesMillen_1972.pdf

I don't mean to deflect the subject of the thread but couldn't help notice on page 8 of the catalog a Mllen 92200 tuner.

I acquired one at a Hamfest. Built like a Tank with a coaxial reflectometer.

Millen made great products.

Phil - AC0OB
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
KK4RF
Member

Offline Offline

Posts: 158


« Reply #17 on: December 22, 2012, 12:33:02 PM »

Season's Greetings to All:
     I have nothing to add to all the above except that this has been an extremely good thread. I've learned a lot about RF chokes which has even induced me to look up more info in old handbooks and QST articles. In the past, when I built something, I always just looked for a junkbox choke with close to the appropriate  inductance and felt like I'd done enough. On one transmitter project, I wound my own chokes on small ferrite toroidal  cores ( the Millen choke no longer available anywhere), and when I could never get the transmitter to work right I always wondered whether the choke substitution might have been the culprit. Now I believe that the chokes were the problem (instability, big time). Makes me want to get out and dust off my old Millen Grid Dip Meter!
     Many thanks to Martin for starting an excellent topic and to all the others  with their input.   ---Marty, KK4RF---
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #18 on: December 22, 2012, 12:57:28 PM »

Yes, the James Millen components were excellent. See them at a fest and at a reasonable price is like finding buried treasure!

Here is a plate choke I wound and overkill for a tube project but it serves as a reference point since it is sort of large. This is a Delron rod 3/4 inch diameter and 10 inches in length. It measures 230 uH (0.23 mH) and has a R value at 400 kHz of 2 ohms. It is SELF resonant (anti or parallel) resonant at 3.7 MHz and by the time you get to 30 MHz it looks a capacitor of 7.8 pF in series with 43 ohms, so a pretty poor Q cap. If you place it on a chassis and short its terminals it has one series resonant point at 16.5 MHz. Otherwise I found no other resonant points. It is wound with about 120 turns of #22 enamel and the winding is broken about a 1/3 of the way up and then continued. Kapton tape was used to hold in place until the terminals were added. For 160-20 M it may be fine. I would be a little concerned about the series R present at the high frequencies. However, it was not intended for high power, much less than 500 W (250 W a couple of 811 or 813 tubes). It would be great to see other measured examples. For example, the Ameritron or Alpha etc... 


* RF plate choke.jpg (61.45 KB, 947x153 - viewed 1891 times.)
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.141 seconds with 19 queries.