Received-SPF: neutral (google.com: 193.82.116.20 is neither permitted nor denied by best guess record for domain of owner-rsgb_lf_group@blacksheep.org)
Date: Sun, 22 Jul 2007 09:22:58 +0100
From: John GM4SLV <gm4slv@sighthound.demon.co.uk>
To: rsgb_lf_group@blacksheep.org
Message-ID: <20070722092258.1e9dfebe@lurcher.twatt.local>
Organization: The Gammy Bird
Mime-Version: 1.0
Subject: LF: 500kHz temporarily QRT
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Status: O


Hello gang,

Thanks for looking for my QRSS last night. Obviously yesterday was not
to be the day!

Had 2 QSOs with Mal 'KEV and a signal report from GM4JYB on Dunnet
Head, so at least some RF is radiating.

I was hoping for a few hours during peak darkness to have try last
night but that was scuppered by my own clumsiness!

My TX uses a VFO running at 8MHz divided by 16. The VFO is stabilized
by a huff/puff circuit and during testing had proved to be rock solid.
I've run QRSS-10 at 15W into test load overnight and the frequency
hasn't varied by 1Hz.

To increase the TX power to that necessary to achieve my ERP target,
which calculation suggests needs 0.7A RF, I fed the PA section with a
variable 0-60V PSU. At 20V I achieved 25W and 0.7A RF. Fine and dandy.

Later in the day I noticed that the VFO was not staying locked when in
TX mode, and once it unlocked it stayed slowly free falling until the
huff/puff circuit was re-set. I found at 10w or so out that it was
stable. I guess there's RF getting in somewhere and upsetting it.

I put in lots of extra decoupling and clamp-on ferrites but to no
avail. The VFO itself is in a screened box of PCB material, but the
huff/puff is open. I'll rebuild the two circuits into one eddystone
box...

During the investigation into the drifting VFO I inadvertantly over
stressed the MOSFETs in the PA (over volts) and turned them into
permanently ON switches.

No spares!!

I was thinking of a redesign anyway, so I'll take this opportunity to
do that. I expect to be back on the air in a day or so. 

Jim 'BMU's design uses IRF520/IRF9520 complementary pair to acheive
push pull from a single phase drive. I think I'll go the old way and
use bi-phase drive and a "normal" pair of N-Channel FETs. I have a
ready supply of IRF150 (or BUZ24) which are TO3 cased HEXFETs that we
use at work in 28 volt 20amp SMPSUs. These are fed with 50V DC and
switch regulate to 28V so I guess they should handle the volts/current
to get 25W at 500kHz? Does anyone have any experience of the devices?

Sorry to have appeared and disappeared so rapidly. That'll teach me to
have enough spares for the PA FETs!

Cheers,

John GM4SLV