Warning: Creating default object from empty value in /homepages/11/d132647312/htdocs/Amfone/mkportal/include/SMF/smf_out.php on line 47
Air inductors for PWM filter




 
The AM Forum
August 09, 2020, 12:58:07 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: Air inductors for PWM filter  (Read 1804 times)
0 Members and 1 Guest are viewing this topic.
N4LTA
Contributing
Member
*
Offline Offline

Posts: 989


« on: January 16, 2020, 03:50:30 PM »

Using Jim Tonnes paper and design model, I came up with a filter design for my 350 watt PDM generator.  I'll use new film capacitors. The inductors I will try are something I have been thinking about for a while. I had been looking at using FT200 -43 or 61 ferrite cores, but these can saturate at some point. I found these air wound inductors about the same size. They are air inductors for speaker crossover use. These are 1.1mH and 1.5 mH but they have a large variety of wire sizes  and inductance. These are from Parts Express. Resistance is less than 1 ohm on each.


* inductors.jpg (91.27 KB, 640x480 - viewed 150 times.)
Logged
KQ6F
Member

Offline Offline

Posts: 87


« Reply #1 on: January 16, 2020, 05:33:25 PM »

I bought some of those too and planned to try them.  But before that I found these:
https://www.mag-inc.com/Media/Magnetics/Datasheets/C058894A2HT19.pdf
https://www.acalbfi.com/uk/Magnetic-components/Cores/Advanced-Powder/p/High-flux-060-PERM-26-9mm-OD/0000001TK4

They have worked out well for me...no sign of any saturation, although my inductor values are in the 50uH range.  Larger values would of course require more turns and might bring the cores closer to saturation.

I should go back and try one of the air cores, especially in the first inductor.

Rod
Logged
M0VRF
Member

Offline Offline

Posts: 103


« Reply #2 on: January 17, 2020, 02:47:15 AM »

Ferrite you quote are for RF use and won't work at the 100's of KHz common with PWM frequencies.

There's much better materials around.

Check out the Super MSS range.

I'm using 1" cores in a 500W modulator and they don't get warm.

Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2497


Bap!


WWW
« Reply #3 on: January 17, 2020, 09:14:14 AM »

If you can find the 58867-A2 core, that one works VERY well for PWM filters for transmitters running about 500 watts or less (DC current of 10A or less).  I have used these in a number of pulse width modulators with very good success.

There is an equivalent core- the CH777060 - works the same.  Another core I've used is the CK777060.  A different material, but very good.

The best I've used, particularly for higher current is the type 2 material from Micrometals.  The T300-2D is a really good core.  I've used it as the INPUT inductor (and all of the others, but the input inductor does most of the work) for 1kW pulse width modulators.  The material is extremely stable over a wide range of flux densities.  The only disadvantage for lower power transmitters is that, due to the relatively low perm of the core, you need a lot of turns to get the higher inductance needed for lower current.

On a 1kw modulator, I used 2 stacked for the input (although 1 would have been just fine), 4 stacked for the middle inductor (highest inductance in the filter), and 3 stacked for the output inductor.

Recently purchased some T400-2 cores for a 10A modulator (45V @ 10A), and the core is big enough to fit all the wire needed to make the inductor.

The *most* important thing when choosing a core is the stability of the permeability with changes in current.  The current in a modulator will go from 0A to 2.5x or 3x the carrier current value (depending on the capabilities of your modulator and the division of carrier DC voltage to power supply voltage).  Figure on at least 2.5x the carrier DC current, and you'll be safe.

Of course, the core must be able to work in the frequency range of the modulator.  My modulators switch at around 160kHz.

So, if you can find something like the 58867A2 or the CH777060, these cores work VERY well with carrier DC current of up to 10A (this assumes a peak current of around 25A under high positive modulation).

Get the free inductor design software from Micrometals and Magnetics.  This will allow you to see what the inductance will be over the current range you need, and the software will help you identify specific parts that you can use.

The idea is to get the most stable inductor possible over the current range in question.  Air core is best of course, but air core takes a lot of space which is why we use cores.
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
N4LTA
Contributing
Member
*
Offline Offline

Posts: 989


« Reply #4 on: January 17, 2020, 10:57:54 AM »

Thanks for the information. I got some from CWS Bytemark coming.

I'll test the air cores this weekend for the small modulator.

Pat
N4LTA
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2497


Bap!


WWW
« Reply #5 on: January 17, 2020, 02:24:22 PM »

Yeah, something like the MS-300060-2 would work nicely for that power range.
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
N4LTA
Contributing
Member
*
Offline Offline

Posts: 989


« Reply #6 on: January 27, 2020, 06:11:59 PM »

First though, for the first transmitter, i will try the air wound inductor. Can't really s a down side as they have no core to saturate.

Pat
N4LTA
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2213


Making AM GREAT Again!


« Reply #7 on: January 31, 2020, 12:55:35 AM »

Pat,

I did the same thing.  Granted, my coils where really big.  Upside, they where easy to right angle.  One coil was at the top, one in the side of the pine sheet.  Downside, they where really big.

In the end, I left them as air wound.  I have coils wound on yellow ferrite that are a LOT smaller, but I like the proven design

--Shane
KD6VXI
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 989


« Reply #8 on: January 31, 2020, 09:32:37 AM »

After  posting above - I found they are really big.  I mounted them at right angles and they are really big. I got some of the recommended ferrite  and they are the right size but permability is too small to get what I need. Before I go further, I need to determine the impedance at the transmitter. Maybe I am thinking too large at 50 ohms. 155 volts at 2.9 amps at 100 percent modulation is my best guess. 50 or so ohms?   The air core 1.0 mH and 1.5mH are big but they fit the task at 50 ohm. I see where Nigel calculated his 350 watt transmitter at 13 ohms -- so am I doing something wrong?

I was assuming  155 volts at 2.9 amps = 53.4 ohms or should this be done at carrier only level???

Thanks for any help or enlightenment.


Pat
N4LTA
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2497


Bap!


WWW
« Reply #9 on: February 05, 2020, 11:11:16 PM »

Ok, when you calculate your filter, first figure out the equivalent resistance of your RF amplifier (at carrier is the easiest).  That is the voltage fed to the RF amplifier divided by the current (all at carrier).

Now you can design your filter.  For lower power transmitters (400 watts and lower - power output), a 4 element filter will work just fine, assuming the corner frequency of the filter is WELL below the PWM switching frequency.

When using an inductor that takes a core, you have to design the inductor to be able to withstand the highest peak current that will be encountered at the maximum output of the modulator.  Using an example, if your modulator has a 120VDC power supply, you can assume that it can actually put out 120V peak.  If your carrier power is - for example - 40VDC at 10A (400 watts DC input), and the modulator can put out 120V on peaks, you have to figure that the modulator will also be delivering 30A of current into the RF amplifier at its peak output.

Your inductor has to be able to withstand 30A peak without a lot of variation in the actual inductance.  The permeability of the core should not drop very much as the DC current goes up.  Most manufacturers of cores supply software that you can use to calculate the performance of DC biased inductors over a wide range of DC current.  The software will usually plot the inductance curve over a range of DC current values.  You want the inductance to be fairly stable over the current range your modulator can develop.
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
N4LTA
Contributing
Member
*
Offline Offline

Posts: 989


« Reply #10 on: February 06, 2020, 09:20:09 AM »

Thanks Steve. That was what I was thinking but was coming up with 40-50 ohms and on Nigels's modulator, he calculated the filter for 13 ohms.

50 ohms gives large values of inductance and I may be stuck with air cores.


Thanks again for your help. I am going to build another of your circuits to be able to vary my pulse duty cycle this weekens.

Pat
N4LTA
Logged
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2497


Bap!


WWW
« Reply #11 on: February 06, 2020, 10:20:07 PM »

Thanks Steve. That was what I was thinking but was coming up with 40-50 ohms and on Nigels's modulator, he calculated the filter for 13 ohms.

50 ohms gives large values of inductance and I may be stuck with air cores.


What is your DC voltage and DC current (going to the RF amplifier) at carrier ?
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
N4LTA
Contributing
Member
*
Offline Offline

Posts: 989


« Reply #12 on: February 15, 2020, 03:15:54 PM »

Steve,

Sorry this took a while but I now have the RF deck working an it looks like the deck input impedance is about 28 ohms. At 60 volts the current was 2.14 amperes and output power was about 120 watts.

Pat
N4LTA
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 0.071 seconds with 18 queries.