[R-390] For your consideration, Imperial Electronics Inc. Receiver restart today
Jim Whartenby
old_radio at aol.com
Sun Jul 19 17:22:22 EDT 2026
Fred
Thanks for the kind words. I choose the 10 volt 5 watt Zener so that the diodes would not need heatsinks. Each diode should only dissipate 2 watts @ 200 mA. If you are concerned, you can always epoxy the diodes to the aluminum chassis.
Audio amplifier should be biased Class A so the volume control setting should make little difference on current consumption.
Enclosed is the power supply annotated schematic with various current measurements.
Regards,
Jim
Logic is a systematic method of coming to the wrong conclusion with confidence. Murphy
On Sunday, July 19, 2026 at 03:54:16 PM CDT, Fred Moore <fred_moore at usa.net> wrote:
I like Jim's idea of 5w Zener diodes in the transformer center tap --
especially because they can be more easily incorporated in the power supply
unit itself, with a little heat sink.
Does anyone know what the total B+ current is in the R-390A with the audio
volumes at zero?
Has anyone measured the B+ current in an R-390?
-Fred
------ Original Message ------
Received: Sun, 19 Jul 2026 07:56:52 AM GMT
From: Jim Whartenby via R-390 <r-390 at mailman.qth.net>
To: "r-390 at mailman.qth.net" <r-390 at mailman.qth.net>
Subject: Re: [R-390] For your consideration, Imperial Electronics Inc.
Receiver restart today
John
Why does this "excess voltage" need to be removed? The R-390A was designed
to operate over the line voltage range of 105 to 125 vac or 115 vac +/-
10%. This is not a regulated power supply so any variation in the line
voltage will affect the B+ and heater voltages directly. Going to a solid
state B+ rectifier will increase B+ by perhaps 20 volts give or take so the
increase in B+ is about 9% over nominal. Since the voltage dropped across
the 220 ohm resistor is current dependent, the voltage regulation of the B+
power supply will be worse with this dropping resistor installed.
It would be better to put two 10 volt, 5 watt Zener diodes between the high
voltage center tap of the power transformer and ground. This is usually used
to generate a negative bias in other radios and is also known as back
biasing. Using this technique, there is no current dependent voltage drop,
there will be a constant 20 volt reduction in the B+ voltage regardless of
power supply current consumption so the B+ voltage will not vary.
Removing the 26Z5 rectifiers will reduce the heater current load on the power
transformer by some 10 watts. This will reduce the stress on this
transformer. Since most of the vacuum tubes in the R-390A are pentodes, a 20
volt higher B+ will have a minimal affect the current drawn from the power
supply. Pentodes are constant current sources, this is why the graphs of
current vs voltage are flat. So their plate current is hardly affected by
changes in the B+ voltage. The few triodes in the R-390A are run at reduced
B+ voltage so the affect on these tubes is also minimal.
The 3FT7 or RT510 current regulator has a voltage range from 8.6 volts to 16.6
volts and a current range from 290 to 330 mA. So one should expect about
12.6 volts nominal across it and about 310 mA nominal thru it. I am not sure
why you would measure 7.1 volts across one of the oscillator tube
heaters.
Perhaps if you measure the heater voltage across both tubes and the RT510 you
should find that they equal the heater voltage winding on the power
transformer. It should be noted that all series heater strings, like those
in AA5 radios also behave as a current regulator. Increasing the line
voltage has a minimal affect on heater string current. There is no linear
relationship between heater voltage and current so the tube heaters act as
their own current regulator.
The current regulator idea was a Signal Corps requirement, Collins thought
that it was not necessary. I don't believe that Collins has used a current
regulator in any of their non-military designs. This information is found in
the R-389, R-390 final report. Obviously, the Signal Corps did not grant a
waver to this requirement when the R-390A was designed.
Regards,
Jim
Logic is a systematic method of coming to the wrong conclusion with
confidence. Murphy
On Saturday, July 18, 2026 at 10:54:34 PM CDT, jkharvie
<jkharvie at verizon.net> wrote:
Jim,First, let me thank you for your generous response. Great to hear from
you againYou raise some interesting points regarding the EIB-895
modification. I looked at some of the logic and think that clearly while yhr
dual tubes works, replacement by solid state diodes provides a more efficient
management of electron flow. Yet, this is efficiency has some excess voltage
that needs to be removed. It is likely that the shift in excess energy to
heat in the additional power resistor really does not totally define a new
operational receiver profile. But I think it is an OK element to take the
time and undertake. I do not see any downside in this.
I am now closely measuring all the voltages on all the tubes in the IF deck
then RF to assess what is and is not taking place.
I see that the designers of the R390A placed the heaters of two valves in
series on the R510 "Current Regulator", these are V505 and V701, I now
measure 7.1 volts on V505, a bit high for a 6.3 design point.
I noted that V507 has a dedicated 4 ohm voltage dropping resistor in the
heater feed to the heater on this valve. If we presume 6.3 VAC source and
0.3 A heater current, we get about 1.2 volt drop, the tube operates with 5.1
VAC. I never noted all these small details when the receiver was
operational.
Regarding the tube, and tube testing. When I removed the tubes the receiver
was working last time. I indexed what tube went into what deck and what
position. The tubes have been maintained in a box inside the home. So, bad
tube - possible, but more likely I introduced a bug, exactly where, I do not
yet know, but I do hope I will find it.
Regarding v501 - yes, my GE tube handbook also indicates max screen and plate
voltage, at 300 and 330 v respectively. So we do have some headroom here.
I will post the results of the tube voltages
thanks for your time and interest
JohnN3JKE
On Saturday, July 18, 2026 at 06:37:05 PM EDT, Jim Whartenby via R-390
<r-390 at mailman.qth.net> wrote:
John
You are being cautious so that is a good thing. You didn't mention that you
had tested the tubes to be sure that poor performance of the R-390A is not due
to weak tubes.
That EIB-895 mod you mention has always bothered me. I have the gut feeling
that it is based on bad data.
To start, the 26Z5 tube manual data indicates that the typical voltage drop is
approximately 35 volts at 200 mA per plate. So with the two plates in
parallel, the nominal tube voltage drop should be around 18 volts since the
internal tube resistance are now halved. I don't see why an 18 volt increase
in B+ would be an issue. A line voltage increase of about 7 volts will do
the same thing.
As for the 6AK6 having a problem with a high B+, I think that there is
something else going on. The tube manual data is misquoted saying that 180
volts is a maximum B+ value. But it is clearly stated in the tube manual
that this is a typical operating voltage. The maximum operating plate and
screen grid voltage is given as 300 vdc. The R-390A maintenance manual
indicates that the screen grid of the 5AK6 is nominal with 215 volts
applied. So I don't think that this is the real problem.
Field Change 1 addresses an oscillation problem with the 6AK6. It changes
how the suppressor grid is wired for V603. It directly connects the
suppressor grid to the cathode instead of to ground. One assumes that this
breaks the feedback path so as to prevent the UHF oscillation. Why V604,
another 6AK6, does not get the same treatment is a mystery.
I would think that better B+ filtering at RF would also break any feedback
paths. Paralleling C603B with a 0.01uF ceramic bypass capacitor would be a
smart thing to do. Might as well do the same for C603A and C603C as well.
Electrolytic capacitors are known to act more as inductors then capacitors at
a few megahertz so any additional filtering at RF frequencies would help keep
things stable.
Comments?
Jim
Logic is a systematic method of coming to the wrong conclusion with
confidence. Murphy
On Saturday, July 18, 2026 at 12:58:41 AM CDT, jkharvie via R-390
<r-390 at mailman.qth.net> wrote:
Greetings After about 10 on and off years, I finally have one of my R390A's
to the point where I plugged it in to check it out. These receivers weigh
more than I remember when they are all assembled!
Following the various manuals along, very helpful to have the multiple
editions of all the terrific community papers, schematics, and manuals. I
found that a number of the resistors I replaced in 1999 needed to be replaced
again.
As I decided that you only live once, this meant I needed to make this my
curtain call receiver rebuild. RF deck, IF deck and the rest of the works.
I slowly powered the receiver up last night, 3 meters active monitoring,
enjoyed looking at the tubes slowly turning on with the lights down low to see
if all started at about the same time, and at a similar brightness. Good so
far. Good smell, no sparks, no arcs, no smoke, no fire. Receiver started and
it was nice.
Nominal voltages and currents on start up using the variac. B+ at about
147.5 v.
Major mods other than new resistors and caps include a move of this unit off
of a dependency on tube diodes onto some solid state diodes that plugged into
the tube sockets, and I did add per the EIB-895 the new power resistor, value
of 220 ohm, used a 10W tucked under the blank-off plate in the AF module in an
aluminum stand-off I made, held down into place with one of the existing
blank-off plate machine screws.
I spent a lot of time and effort with the IF deck, one in and one out and this
deck now has zero carbon composition resistors.
So...the receiver is now back together, mechanically aligned, cams are in good
placement and tracking, RF switch appears to be aligned, crystal deck appears
to be aligned, 10-turn stops in place, all smooth and continuous and proper
rotation etc.
Hooked up speaker to the unit via a small xfmr that was 120v in 12v out as a
good enough audio impedance match for the 600 ohm to 8 ohm speaker
Electrically the following demonstrate functions that resemble alive and maybe
well, I picked up a Direct Digital Synthesizer electronic signal generator as
a source of frequency, have a 4-channel Tek TAS-475 scope that I had to
rebuild the switching power supply to get it going, but it is now operational,
as is my trusted Triplet 630 meter and my newer DMM's.
In general the receiver appears to be on the way to be functional...but needs
some work...Following is OK:
Power SupplyLampsElectrical Bias so far as measured, good
Tube v501 in IF deck has measured good electrical characteristics using DMM on
Plate of 183.1, Screen 97.3, Cathode 5.4, Grid ~0.1 v DC
Mechanical Band pass filters check out
Detector circuit lives
AGC - vice MGC - TBD
Wide narrow filter - checks out
Meters - both Line and Carrier are operational
BFO - works
RF gain - appears to be working (some stage lack of gain concerns)
Local Audio Gain - appears to be working
The key issues to resolve1) Calibration onboard not yet working, nothing
here yet
2) Gain in IF and RF does not appear to be where it needs to be
I intend to identify and fix why these are not working
Using E211 on the RF deck, it now takes more signal injection than should be
needed to hear on the speakerWhen I plugged a 3 foot long wire into E211,
interesting that the receiver received many spurious emissions from the signal
generator, these were picked up and amplified, this was encouraging.
I ran a couple of tests measuring DC bias on diode detector with signal
input+BFO at levels of:Input Diode Load Measurement
0.002v = -0.509 volts
0.003v = -0.530 volts0.005v = -0.582 volts0.006v = -0.632 voltsThe
outcome shows a gradual increase in negative bias but not at the gain
sensitivity level indicated in the manual, i.e. -7.0 volts with 150mv of
455kHz applied to the IF Deck
With the BFO on and input of 455kHz at 0.04 volts into E211 measured -12.5DC v
at the diode output (rear terminal strip)
Does the group have knowledge on the specific response per stage of the IF and
the RF sections for applied signal gain at each tube?
What did I miss? As I read the manuals and look for specifics to resolve.
thanks in advance for the words of wisdom.
JohnN3JKE
______________________________________________________________
R-390 mailing list
Home: http://mailman.qth.net/mailman/listinfo/r-390
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:R-390 at mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
______________________________________________________________
R-390 mailing list
Home: http://mailman.qth.net/mailman/listinfo/r-390
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:R-390 at mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
______________________________________________________________
R-390 mailing list
Home: http://mailman.qth.net/mailman/listinfo/r-390
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:R-390 at mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: R-390A B+ power distribution.pdf
Type: application/pdf
Size: 50175 bytes
Desc: not available
URL: <http://mailman.qth.net/pipermail/r-390/attachments/20260719/82eaa8f8/attachment.pdf>
More information about the R-390
mailing list