Re: LF: Poor aerial connection affecting gate waveforms??

[Andy Talbot <andy.g4jnt@gmail.com>]

I don't know why more designs based on them haven't appeared.  Other than my half bridge version running from 320V DC from raw rectified mains, all the other high power amps around seem to use the classic pair of devices into a push-pull transformer. With all the baggage such designs entail.

50V supply,   full H-bridge, tank on the output from the bridge, single DC isolation / impedance matching transformer.    No inductors other than the  (ideally air cored ) one in the tank.    Can use the Decca guard circuity, which beats hands-down all these trips and overcurrent and drive failure arrangements.

The advantage of doing one from a 50 - 100V rail over my direct mains one is that drive to the FETs doesn't have to go though an isolation transformer.  You can get complete single-chip H-bridge driver chips that handle the bootstrapped supply and just need a logic level drive signal.

Andy  G4JNT

On 15 April 2017 at 17:27, Graham <g8fzk@g8fzk.fsnet.co.uk> wrote:

Can only go  by the  images  in the links  I sent a while back .. I was wondering,  after  Andy's  last post ,

Why not  simply  build a  copy  of the   Decca  out put  stage ,  series tuned cct's  and the  3 -2-1  adding  transformer
there  must be enough  of them  about,  for the  transformers  and chokes to be  identified , the  same  or  similar must be  about ..
tried and tested configuration ?

Taking  everything , via  the  , what appears to  be , a step  up , transformer , will  provide  isolation  from  the  Ae  , if you combine  at 50 ohms , then  everything  will  pass  back to the  amps , via the combiner ,

Could be  totally  wrong  of course, but there must of  been a  reason  for the  lay out ,  when  other  multi unit  pa's  used  50 ohm  combining networks
but its  tried and  tested and  what's better  than  2   500w power modules ? .. 3 of course ,)

73-G,





--------------------------------------------------
From: "Chris Wilson" <dead.fets@gmail.com>
Sent: Saturday, April 15, 2017 3:10 PM
To: "Graham" <rsgb_lf_group@blacksheep.org>
Subject: Re: LF: Poor aerial connection affecting gate waveforms??

Hello Graham,

Saturday, April 15, 2017, 2:14:17 PM, you wrote:

Chris,

I assume , due to the  Q any small  variation , will  move  the
tune  point  about , but  looking  in the   data base , Signals ,
better last night  to  sv8 than  the   previous night , and  may be
up  1 db at pa7  , not to say  the  noise was less , but  at  -10 db
that level   will  be  much  higher  than the  local noise level ,
should be a  reliable  indication

How important  are the  wave forms ?

73-G,

I don't know how important they are I am not a FET ;) All I know is
the output transformer and the LP filter bank central toroid run a lot
cooler if the wave forms aren't all spiky. I hope someone can say how
critical this is? Nothing went pop, so presumably not ultra critical,
but heat = lost power I would think? Cheers G!


--
Best regards,
Chris                            mailto:dead_fets@gmail.com

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus