[Amps] Drake L-4B Should I or Shouldn't I?
Jim Thomson
jim.thom at telus.net
Mon Mar 31 06:18:30 EDT 2014
Date: Sat, 29 Mar 2014 18:12:00 -0700
From: Don Jones <ko7i at comcast.net>
To: amps at contesting.com
Subject: [Amps] Drake L-4B Should I or Shouldn't I?
Last year I picked up a L-4B from an estate sale. About a month ago I
tried to fire it up. Looks like I got some problems to repair first. I
did hear a thump come from the transformer of the L-4PS as it charged
the high voltage cap bank. No biggie, also I did see the 3-500's light
up in the RF deck.
What I did not see was any indication of High Voltage being present on
the L-4B's voltage meter. Fingers crossed I do not have a blown meter.
What should I see for pin to pin resistance of the High Voltage meter?
Also what resistance should I see for the plate current meter? This has
me puzzled, I would think that the meter should have shown that high
voltage was present. I presume that this is a proper assumption. Correct?
Since I suspected there was problem, I stopped there and did not attempt
to "key" the amplifier and see if it would tune up. I pulled the 240VAC
power plug and let it set for a few days. After which I opened the high
voltage power supply section and inspected it internally. The power
supply has the original paper caps that came with it from the factory. I
did not see any blown diodes or smoked resistors, the paper caps still
had their guts which is a good sign.
After some thought and having good service from Harbach Electronics with
a Clipperton-L power supply rebuild kit I decided to order one for my
L-4PS so I could get those old caps and diodes replaced. I also ordered
a new rebuild kit for the front panel power switches and a soft start
kit for the amplifier too.
As for the Should I or shouldn't I question...
I am going to operate this amplifier with an out board QSK switch and
bias control. I am thinking about hard grounding the grids and
installing a 50k ohm resistor in the B- lead to cut off the tube instead
of relying on the +150Vdc (or so) cathode voltage as designed from the
factory. I have heard "internet stories" claiming that the L-4B can
become unstable if the grids are hard grounded. Something about the gain
of the tubes becoming higher and unstable. Is this true or not? I do
want to use this amplifier on all bands including 10 meters.
My bottom line goal is to run this amplifier at about 750 to 800 W rms
out and enjoy it for many years to come. Looking for some technically
sound advice and objective criticism. Sorry for being long winded. Thanks!
73, Don KO7i
## I own FOUR drake L4B amps. You can get the complete combo diode and
8 lytic board from http://www.theheathkitshop.com/page24/index.html. It takes up the
same room as just 4 of the oem drake caps. Room to install 2 such boards if you want..so u can
double the total uf. You can buy just the board too.
## I tossed the mickey mouse semi floating 200 pf grid caps + the pair of grid chokes.
Bond all 3 grid pins to chassis with .25 to .5 inch wide cu strap..and be done with it. And no, it
WONT make the amp unstable. It only has 2650 vdc on the tubes on RX..and 2500 vdc under
load of 800 ma. Drive power requirements will drop 20-25 watts. Since the xcvr is then running
lower po, the imd from the xcvr will be a lot cleaner. IMD will drop, when measured at the output of the
amp.
## Dump the +130 vdc cut off bias fubar nonsense. BTW, its +90 vdc cut off bias when on CW position with 1900 vdc.
## use a pair of paralleled 100 kw, 3 watt mof resistors for cut off bias. Then re-wire the center contact of the 3pdt oem relay
so the 50k is shorted out on TX. Drake used zero bias on the L4B..which results in 220-230 ma of idle current....more fubar.
Use 10 x 1N5408 or better yet, 6A10 diodes in series for bias. Then u end up with 100 ma idle current on ssb..and 40ma
on CW position. For rock solid bias, install a 30,000uf to 50,000 uf lytic across the string of diodes. I used a simple spdt-
center off toggle on rear apron to provide 3 x different bias voltages. 3 diodes, 7 diodes, or all 10 diodes.
# throw away the pair of 100k, 50 watt wire wound bleeder resistors in the power supply, you don’t need em. You already
have 100 kw 3watt mof EQ resistor across each lytic anyway. And you wont see the B+ increase at all without the huge bleeders.
Ok now the HV cab will run barely luke warm instead of blazing hot. Ditto with top cabinet of L4B RF deck. I then installed
a 50 ohm- 50 watt wire wound resistor in series with the B+. You can install it either in the RF deck, at the HV connector...or
in the power supply cabinet..where the huge bleeders once resided. You can also use a pair of 100 ohm- 50watt wire wounds..
in parallel to make the 50 ohm glitch assy. The .82 ohm HV fuse resistor remains..and should be in front of the 50 ohm glitch assy.
## I added external step start..with a 25 ohm, 50-100 watt metal finned, bolt down resistor. A P+B 30 amp DPST relay applies the
240 vac. A 2nd 30 A SPST relay shorts out the 25 ohm step start resistor. Then u get to step start the HV caps AND the fils of the
tubes..at the same time.
## Put RVS connected 1N5408 or 6A10 resistors between B- and chassis. Then you will never blow out the plate or grid meter.
## If u want QSK... use a gigavac GH1 vac relay setup. I used 3 of em. 1 for the rf input, 1 for the rf output..and the 3rd one shorts out the
50 k cut off bias. Then u can stuff 2 kw though the RF deck, when the RF deck is on bypass.
## I string all 4 of my amps in series, nose to tail, all tuned up on different bands. A simple rotary switch ensures that only 1 amp can ever be keyed
by the xcvr at any given time...simple interlock Rotary switch has the xcvr key line on the input... and one output per amp.
## adding 160m is simple too. A pair of padded arco caps + a small torroid make up the tuned input...which is relay switched inline.
The plate choke is already 154 uh, so u don’t need to touch it...... just add a pair of 4700 pf bypass caps to the base of the choke.
The bifilar is aprx 28 uh...so u don’t have to touch that either. Use AG6K 160m mod to simultaneously pad both the tune and load caps..
and switch in the 3 x T2252A torroids with 10 ga wire.
later... Jim VE7RF
More information about the Amps
mailing list