The AM Forum

THE AM BULLETIN BOARD => Class E Forum => Topic started by: W4AMV on November 16, 2022, 01:12:43 PM



Title: Class D paper by Raab and device sources
Post by: W4AMV on November 16, 2022, 01:12:43 PM
Good day,

I was reading a paper by Fred Raab and considering looking at his work on class D.

The devices in question are the drivers EL7144C and the APT ARF440 and 441. Any known sources
or reasonable equivalents? Thanks in advance, Alan


Title: Re: Class D paper by Raab and device sources
Post by: KD6VXI on November 16, 2022, 06:30:35 PM
I've built amps and rf decks out of ARF devices.

They are pretty much unobtanium.

I had a source of thr ARF448A and B siamese pairs, but that dried up maybe 6 years ago.....?  It was a dismantler in China. 

You'd do better looking at the Mrf300 and similar devices nowadays.

--Shane
WP2ASS / ex KD6VXI


Title: Re: Class D paper by Raab and device sources
Post by: W4AMV on November 17, 2022, 08:43:32 AM
Thank you Shane. The 300 is a bit pricey. I have some Cree SiC devices. They might fill the slot if the Rds on and Cin, Cout are decent. I could play with the layout to get the FET orientation reasonable.

Any thoughts on the driver from your experience?

Tnx, Alan


Title: Re: Class D paper by Raab and device sources
Post by: W4AMV on November 17, 2022, 12:53:02 PM
The SiC that I have available are,

C2M0080120D
C2M0280120D
C2M1000170D

The closest to the ARF I believe in hitting Rds on and C values would be the C2M028xxxx as the others have either lower C but larger Rds on. Achieve a lower Rds on comes with larger C in/out but the C2M028 provides a decent tradeoff.

Thoughts?


Title: Re: Class D paper by Raab and device sources
Post by: KQ6F on November 17, 2022, 07:32:47 PM
When choosing between SiC devices I think it depends upon the circuit topology.  I saw the Raab paper years ago but don't remember its topology.  In 2016 I built a rig using four C2M0160120D devices.  It has been perking along ever since without a single hiccup.  They are rugged devices.

I'm using two IXDD614CI drivers to drive two transistors.  So the cap load on each is a little over 1200pf.  I have each driver mounted onto a very small heatsink which is probably not needed.  I can detect no temperature rise, even after long transmissions.

The four SiC devices are also mounted onto a very modest heatsink and I can't detect any significant heating there either.
But the Rd(ON) for these is only 160 m ohms each

So, bottom line, I think your choice comes down to your driver capability to drive capacitive loads and the size of the heatsink for the SiC transistors.


Title: Re: Class D paper by Raab and device sources
Post by: W4AMV on November 17, 2022, 08:36:59 PM
Thanks.

Raab's work was in Communications Quarterly and is push pull. The Elantec EL7144 is used as a gate driver and they are driven to obtain PP drive. No input transformer used. The FETs in use 730 pF g-s and g-d 77 pF. The SiC devices I have are well within that range. So a driver like the 7144 should be ok. I need to find something comparable say from IXYS...


Title: Re: Class D paper by Raab and device sources
Post by: W4AMV on November 22, 2022, 09:43:12 AM
This is the same paper as appeared in the Quarterly. See APT application note APT9403. There are additional app notes that provide good design info.


Title: Re: Class D paper by Raab and device sources
Post by: WBear2GCR on November 28, 2022, 08:27:43 AM
That's 30 year old (approx) info...
afaik there has been a lot of work and products made using Class E in the intervening time.
Also new(er) devices and drivers...

                _-_-
AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands