Message-ID: <5C0692E5.6060807@posteo.de>
Date: Tue, 04 Dec 2018 15:44:53 +0100
From: DK7FC <selberdenken@posteo.de>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
References: <c0a5c460-47dd-1671-28fb-44590355a877@n1bug.com> <2133457080.64533.1543584595217@mail2.virginmedia.com> <27270b29-dde1-1207-1158-becc1e4de123@n1bug.com> <1382885588.67051.1543589445229@mail2.virginmedia.com> <530677a9-52b1-0280-128f-4afc58f94536@n1bug.com> <CAA8k23SLRjnpDk-=4V_26kWObL=tC9RXNwVp__cxCUd+hcYwmg@mail.gmail.com> <trinity-cc8c170d-cca4-471f-9cfa-4b08ae2e53f0-1543608209738@3c-app-webde-bap21> <CAA8k23Q20B6z9VV_XQxAk8yQw6LAd44XNwhbXUhHrWXJmenJkw@mail.gmail.com> <911f1e76-a994-c5d2-ef25-c4f7fde7d46f@n1bug.com> <5C040AA2.2070003@posteo.de> <5710527747.20181202170155@gmail.com>
In-Reply-To: <5710527747.20181202170155@gmail.com>
Subject: Re: [Lowfer] LF: Re: FET change, hmmm
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Content-Transfer-Encoding: 8bit

Hi Chris,

Am 02.12.2018 18:01, schrieb Chris Wilson:
> Any chance of a few photos of the H Bridge amp construction please?
OK, here are some impressions ;-)
Outside: 
http://www.iup.uni-heidelberg.de/schaefer_vlf/LF/pa/20181204_151804.jpg
Filters: 
http://www.iup.uni-heidelberg.de/schaefer_vlf/LF/pa/20181204_151945.jpg
FETs mounted to heat sink: 
http://www.iup.uni-heidelberg.de/schaefer_vlf/LF/pa/20181204_152015.jpg
Gate transformer: 
http://www.iup.uni-heidelberg.de/schaefer_vlf/LF/pa/20181204_152050.jpg
Some PTT watchdog stuff: 
http://www.iup.uni-heidelberg.de/schaefer_vlf/LF/pa/20181204_152158.jpg
>   I take it the BF862's in the optocoupler feeds act as current limiters?
>    
Yes. I had the idea just to reduce the dependency of the duty cycle from 
the supply voltage level and shape of the driver a bit.
> Any reason you didn't just use resistors instead? Or is it for
> galvanic isolation?
Yes, galvanic isolation. One aspect is the safety (oh yes, i like safety 
but only when it is useful to consider), the other aspect is the 
prevention of EMC issues in the shack. This way, the PC or other 
sensible signal generators is carefully decoupled from the power side.

>   The IXDD414CI appears obsolete, the IXDD614CI is
> still current from none Chinese suppliers, should I use a 614?
Yes, there seems to be no difference except the name.
I got them from https://www.digikey.com/products/en?keywords=ixdd614ci
>   Any
> thoughts on more modern FET's as the IRFP460A is also off the current
> device listings from named suppliers :( Thanks!
>    
Oh! At least here they are available: 
https://www.reichelt.de/mosfet-n-ch-500v-20a-280w-to-247-irfp-460a-p118998.html?&trstct=pos_0
Thanks for the note. I should order a bunch of them as long as they are 
available.
For LF, the normal IRFP460 worked reasonable too but at MF you already 
notice a significant difference in upwarming due to higher switching losses!
Try to get them from somewhere. It should be possible. Or search another 
FET with similar parameters.
RS allows a good parametric search, e.g. 
https://de.rs-online.com/web/c/halbleiter/diskrete-bauteile/mosfet/?sra=p&applied-dimensions=4294424079,4294424036,4294341069,4294374353,4294393595,4294386882,4294421462,4291196625,4294392327,4294407012,4293040733,4294389435,4294337998,4293043627,4291658865,4294421649,4292771966,4294416407,4294215551,4294324063,4294136743,4294325511,4292899339,4294220644,4294204641,4294326243,4294095386,4294303277,4294136519,4293253351,4294324990,4294124426,4294222002,4294136239,4294124482,4294325988,4294222053,4294324870,4294133462,4294137170,4294325552,4294325005,4294324039,4291653797,4294325177,4294325265,4294325021,4294325269,4294325291,4294325157,4294099290,4294136648,4293272063,4294136333,4291453942,4294098433,4294324788,4291658391,4294136487,4294330306,4294324584,4294325413,4294323450,4294136639,4294119632,4294132386,4291653453,4294325018,4294086943,4293245047,4294221031,4293048710,4294116300,4294126751,4291653865,4291453937,4294324581,4292940441,4293063774,4294200481,4294131612,4291653452,4294116869,4294332397,4294086951,4294126698,4292801591,4294136028,4294325168,4294226681,4294095381,4294325172,4294136903,4294323509,4294086948,4291653818,4294087757,4291658390,4294193644,4291653837,4294325565,4291653538,4294116716,4294136800,4294421967,4294422427,4294424078,4294137854,4294413678,4294136280,4294421650,4294127968,4294132407,4294420404,4294409034,4294419000,4294421459,4294421386,4294421369,4294410955,4294128048,4294410089,4294418957,4294421504,4293247874,4294419079,4294403642,4294419075,4294409642,4294416416,4294421582,4294397063,4294422422,4293250149,4294398715,4294421360

> Can you upload screen shots of the
> gate and drain waveforms you are now seeing for comparison, they are
> kind of hard to envisage in words ;)
Unfortunately i did not document this very well in 2012. My camera and 
also the scope didn't really make impressive photos during that time ;-)
I remember the curves were quite smooth and there was a transient swing 
on the positive edge, maybe up to 140 % and then back to 100%. This 
transient swing is smoothed with the gate resistor, on the cost of a 
slightly lower rise time.

Regarding the 2 kW statement: You know this always depends on the 
temperature of your heat sink and thus, the strength of the fan and the 
ambient temperature in the shack. I never run it at 2 kW but something 
near 1.5, during winter.
My fan is stepless regulated in speed (not like the annoying 
on/off/on/off switched fans in most commercial power supplies), so the 
speed gives a feedback about the heat sinkl temperature. In winter, at 
high power, it was running quite relaxed mostly, even at WSPR-30! :-)

Hope this helps...

73, Stefan