Daily GLOWBUGS
Digest: V1 #51
via AB4EL Web Digests @ SunSITE
Purpose: building and operating vacuum tube-based QRP rigs
Subject: glowbugs V1 #51
glowbugs Thursday, June 5 1997 Volume 01 : Number 051
Date: Tue, 3 Jun 1997 20:20:24 -0700 (PDT)
From: John Kolb <jlkolb@cts.com>
Subject: Re: High voltage in a PC power supply (was: Re: Electrolytic caps. needed...)
On Tue, 3 Jun 1997, Murray Kelly wrote:
> I wonder if one could put the output (5V) into another
> SMPS transformer back to back, thus creating an isolation
> type trf? You'd need another old SMPS, of course to
> supply the 2nd trf.
> That way you'd get the full 300V and not run the risk
> of fiddling with live chassis?
You mean connecting the 5 V AC, 50 kHz square wave directly
to the LV secondary winding of a near identical transformer,
right? Should work pretty well. I hate to work on switching power
supplies without a proper schematic, and that's very hard to
get for computer parts.
>
> How about modulating the 5V, even?
>
Probably a little tricky to modulate the 5V AC, but the switching
regulator circuit should have a reference input, that you could
transformer couple audio into to modulate.
John Kolb KK6IL
Date: Tue, 03 Jun 1997 23:01:54 +0100
From: BOB DUCKWORTH <bob@atl.org>
Subject: Another neat RX.
No I didn't build it but am very happily working on someone elses
HBR-16. The Home Brew Receiver 16 tubes article from the Oct '59 QST
appears to have gotten someone so fired up that he had it built before
the Dec QST and the 5763 product detector mod.
If the Mosley CM-1 description bored you, stop reading now.
The Mosley had 5 tubes (actually 10 as there was a triode and pentode
in every one of them).
This HBR-16 has 16 tubes (actually 17 if you count the one 12AU7, 2 x
triode as two).
I don't know about RX design so I'm gonna ask some questions and
hopefully some will be kind enough to post what they know so I can get
this thing running in top form.
Circuit starts with a 6BJ6 Pentode as RF amp. Plug in coils for each
band. Antenna link coupled to grid coil which is tuned by one of three
sections of the 5-23mmf/section VFO cap. RF gain via cathode bias pot.
Mosley manual said triode amp is better on high bands owing to
noise considerations. Real feature or just making what they
used look good?
Output of the RF amp is coupled the same way (link to tuned grid) of
the first mixer 6CB5. Plug in coils are used here too with tuning by
another section of the VFO cap.
Any suggestions as to least aggrevating approach to making
sure all these plug in coil based tuned circuits track well?
The VFO, a 6BH6 is tuned by the third section of the VFO cap and also
uses a plug in coil. VFO feeds 1st mixer via same grid as RF amp.
How to determine best VFO injection level?
Output of first mixer is 1600kc and makes it's way to the second mixer
6BC5 via a link coupled, double tuned, transformer wound on a pill
bottle.
Here I'm curious as to how to determine whether ot not the
amount of coupling I have is right, too much, or not enough.
Would a 1600kc xtal (there is one in the junk box) be too
tight as a filter or attenuate the signal too much at this stage
of the RX?
The second mixer gets its' LO from a 1700KC 6C4 xtal osc. and is sampled
through a 100kc double tuned IF transformer (Miller 612-M1) by the 6BJ6
first IF amp. Mixer gain via cathode bias pot.
Again, how to decide what injection level for 1700Kc signal
is best.
Deja-vu, the second IF amp which is followed by the third of the triplet
of transformers. The newly shaped IF signal is then routed to either a
6BE6 product detector or 1/2 of a 12AU7 double triode as envelope
detector. Both IF gains are via one cathode bias pot.
As an operator, what is a good way to tell how much IF gain
is enough.
BFO comes from a 6BH6 with Miller 012-M5 BFO coil and running as
electron coupled osc.
Injection level question again.
The other half of that 12AU7 samples the output of the IF string and
controls gain of the two 2nd IF stages for AVC and also drives the
6CB6 'S' meter amp.
Audio from either of the two detectors is amplified by a 6C4 driving
headphones of a 6AQ5.
Hey, that's only 15 tubes! Well there is also a 6BH6 xtal osc running
all by it's lonesome to provide a cal. signal.
So there. A bunch of gain distribution and injection level
questions about a 37 year old radio. I'm all ears.
- -bob
Date: Tue, 3 Jun 1997 22:52:53 -0500
From: bill@skeeter.frco.com (William Hawkins)
Subject: outer limits of 12AX7
Anybody out there with a working Tektronix 570 curve tracer for tubes?
There's been a discussion of tube ratings following the discovery that
some handbook said a 12AX7 was good for 7.5 watts dissipation. Most
other books disagree, but the 12AU7 is good for 5+ watts with both
sections dissipating half that.
Several people said you must not go beyond the limits, but I wonder
what happens when you do. Suppose reduced life was acceptable. Now
what are the limits? What kind of current can you pass without melting
something, or going too non-linear? Does the hot tube degrade rapidly
from outgassing? After all, there are all these stories about getting
high power from an oil-cooled metal 6L6.
Hey, I've never run a 12AU7 with color in the plates. I've got no idea
what happens. I suggested the 570 because that'd give the facts.
Regards,
Bill Hawkins bill@skeeter.frco.com
Date: Tue, 3 Jun 1997 21:28:54 -0700 (PDT)
From: Ken Gordon <keng@uidaho.edu>
Subject: Re: outer limits of 12AX7
On Tue, 3 Jun 1997, William Hawkins wrote:
> Anybody out there with a working Tektronix 570 curve tracer for tubes?
> There's been a discussion of tube ratings following the discovery that
> some handbook said a 12AX7 was good for 7.5 watts dissipation.
NO! Not dissipation:...OUTPUT!...in class B push-pull.
> Most
> other books disagree, but the 12AU7 is good for 5+ watts with both
> sections dissipating half that.
>
And the 12BH7 has about 3 times the plate area of the 12AX7. I looked
last night. Further, all the tubes discussed here have the same base
diagram so that one could wire up an amp as I discussed here a few days
ago, and plug them all in one at a time and test them...which I intend to
do.
Ken W7EKB
Date: Sun, 1 Jun 1997 22:42:00 GMT
From: ralph.hartwell@emachine.com (Ralph Hartwell)
Subject: Electrolytic caps. needed
K>It looks like I am going to have to replace all of the electrolytic
K>caps. in the power-supply circuits of my children's HW-16 and would
K>I am considering buying some of those photoflash type caps., but am
K>wondering how they would react to continuous use at full working
K>voltage. Does anyone here have experience along these lines?
Ken,
As I recall, many flash caps do not have good ESR values, and may
possibly overheat or explode when used in filter circuits due to the AC
component of the applied voltage. Additionally, because of the poor
ESR, they may not eliminate the hum very well.
Ralph
W5JGV
- ---
þ QMPro 1.52 þ Religious Error: (A)tone (R)epent (B)lame Satan
Date: Wed, 04 Jun 1997 22:39:16 +1000
From: Murray Kelly <mkelly@faraday.dialix.com.au>
Subject: Re: High voltage in a PC power supply (was: Re: Electrolytic caps. needed...)
Mmmmm. The 5V is rectfied and smoothed normally, then fed back
to the controler chip as a sample.
I have a circuit here of *one* SMPS and I have used it as a
reference once. I was able to determine that a rectifier
and the chip had *gone* and replaced the diode with a 4007
even tho it's not high speed, and the chip I fluked at a local
supply house. It worked.
One would still have to sample and smooth a little of the
output for feedback. One would modulate that.
200W (and up) supplies are there for the asking.
I think it would make a pretty neat regulated screen supply.
John Kolb wrote:
> You mean connecting the 5 V AC, 50 kHz square wave directly
> to the LV secondary winding of a near identical transformer,
> right? Should work pretty well. I hate to work on switching power
> supplies without a proper schematic, and that's very hard to
> get for computer parts.
> > How about modulating the 5V, even?
> >
> Probably a little tricky to modulate the 5V AC, but the switching
> regulator circuit should have a reference input, that you could
> transformer couple audio into to modulate.
>
> John Kolb KK6IL
- --
******************************************************************
* Murray Kelly vk4aok mkelly@faraday.dialix.com.au *
* 29 Molonga Ter. / Graceville/ QLD. 4075/ Australia *
* ph/fax Intl+ 61 7 3379 3307 *
******************************************************************
Date: Wed, 4 Jun 1997 08:46:33 -0400 (EDT)
From: EWoodman@aol.com
Subject: Lorenz Type Coils
Shane,
I saw you mention that you have used some Lorenz-type coils in your
transmitter. I've always liked the looks of those. Do the standard coil
winding formulas work for this type of coil?
73 Eric KA1YRV
Date: Wed, 4 Jun 1997 09:23:23 +0000
From: "Brian Carling, Radio AF4K" <bry@mnsinc.com>
Subject: Re: 6L6 voltage
Conard - how many WATTS did you get out of the 6L6 glowbug with 500
Volts on the plate?
Just curious!
Bry
****************************************************
*** 73 from Radio AF4K/G3XLQ Gaithersburg, MD USA *
** E-mail to: bry@mnsinc.com *
*** See the interesting ham radio resources at: *
** http://www.mnsinc.com/bry/ *
****************************************************
Date: Wed, 4 Jun 1997 09:23:24 +0000
From: "Brian Carling, Radio AF4K" <bry@mnsinc.com>
Subject: Re: Electrolytic caps. needed
On 1 Jun 97 at 22:42, Ralph Hartwell spoke about Electrolytic caps.
needed and said:
> K>It looks like I am going to have to replace all of the
> electrolytic K>caps. in the power-supply circuits of my children's
> HW-16 and would
>
> K>I am considering buying some of those photoflash type caps., but
> am K>wondering how they would react to continuous use at full
> working K>voltage. Does anyone here have experience along these
> lines?
>
> Ken,
>
> As I recall, many flash caps do not have good ESR values, and may
> possibly overheat or explode when used in filter circuits due to the
> AC component of the applied voltage. Additionally, because of the
> poor ESR, they may not eliminate the hum very well.
>
> Ralph
> W5JGV
E.S.R.? You gonna have to help us uninitiated ones here Ralph!
Let's see... Extra Simple Radio?
Every Schematic Reeks?
Electrolytic Squeaking Ratio?
Extra Strong Radials?
(grin)
For more jokes see:
http://www.mnsinc.com/bry/jokes.htm
Date: Wed, 4 Jun 1997 09:31:20 -0600
From: dfrancis@iex.net (Dexter Francis)
Subject: Computer Power Supplies
I mentioned this a few weeks ago, but this seems like a good time to
re-state it:
Many computer supplies are switch mode regulators that double or triple the line
voltage before running it thru the chopper/modulatior section. This means that
there is probably a place to get 240 to 360 volts out of these little fellows,
in addition to the +12 and +5. A little fiddling with the +5 regulator could
probably make it yield +6 and voila! an integrated HV and LV power supply for
your next gassless valve project.
- -df
* CWest Tube Sales - P.O.B. 22443 SLC, UT 84122 *
* http://www.usa.net/~dfrancis/CWest_Tube_Sales.html *
* e-mail to: tubes@usa.net *
Date: Wed, 4 Jun 1997 11:22:59 -0600
From: mack@mails.imed.com (Ray Mack)
Subject: Re: Another neat RX. (long)
Here are answers to some of Bob Duckworth's questions:
1) I am doing this from very old memory, but here is the triode vs
pentode story. Pentodes are a lot more stable due to the shielding of
the extra grids. At the low frequencies you don't have to neutralize
and kill VHF parasitics. As you go up in frequency, the triode is
easier to tame because the problem frequency range is now the range
you are trying to use. As to noise, each of those grids is a source
of noise due to electron bombardment and secondary emmision. The
triode only has one grid, so there is less noise.
2) The best way to describe how to get the best injection is to throw
some theory at you. *All* mixers are non-linear devices. They use
the squared, cubed, fourth power, etc. terms of the active device to
multiply one sine wave value by another sine wave value. The best
mixers have the squared term as the dominant term by a factor of 10.
JFET's are terrific mixers because for practical purposes they *only*
have a squared term in the transfer function!
I pulled out the 1977 ARRL Handbook and turned to the vacuum tube
section. It looks like bias and plate load resistance are more of a
limiting factor than injection level. From the curves, the lower the
plate load resistance, the more non-linear the tube will be. You also
want to bias the mixer near cutoff to get the most non-linear portion
of the curve.
That covers the static circuit design. Now we need to look at the
injection level and how it relates to the plate load and bias. Let's
look at the JFET as an example. You have probably noticed that the
idea that having *only* a squared term *guarantees* distortion from a
JFET RF amplifier. This is correct, but if you bias higher up the
curve and keep the signal swing small you have a very close
approximation to a straight line.
Here is the trade-off on injection level: the more non-linear the
transfer curve the less injection you need. and vice-versa. This
effects conversion gain. The less non-linear, the less IF output you
get with more feedthrough of the oscillator and RF which you must then
filter out.
I don't have my tube manual here at work, so I can't give you some good values
for the bias, load and injection for these particular tubes. I'll try to look
tonight.
3) The coupling at 1600 KC will determine whether you have a single hump or
double hump in the response. Over coupling will cause the double hump. You
don't want a single crystal at 1600 KC unless you are only interested in CW.
Even at this high frequency a single crystal will give a pretty narrow filter.
You can try this with a Heath HR-10B by taking out one crystal and substituting
a phasing capacitor. This IF basically provides a rudimentary roofing filter
and image protection. Its only purpose is to make the response as big as a
barn door instead of as broad as a barn. You probably want to shoot for about 6
KC to 10 KC of response at this stage if you want to listen to AM broadcasts.
If you want Hi-Fi AM you might struggle to make it 20 KC wide by didling with
coupling and tuning of the 2 circuits of the transformer.
4) I hope I read this question correctly. The short answer is to put most
(3/4??) of the gain at the low IF and as little as possible ahead of it. You
really don't need any gain at the RF stage and 1st mixer for HF. The rest of
the gain goes at audio.
The textbook answer to how much IF gain is necessary is "it depends". The place
to start is back a little farther. You need to take the size of the minimum
discernable signal (related to the noise figure and worst case noise at the
antenna terminals) and figure out how much power you want at the speaker. The
difference is the total gain needed in the system. There have been several good
articles over the years in QEX and Ham Radio on gain distribution in a receiver.
Uhlich Rohde wrote nost of them. Even the latest versions of the ARRL handbook
do a fairly good job on this subject.
A thumbnail sketch looks like this:
a) put no more than about 20 dB of gain ahead of the narrowest filter. This
keeps signals within the passband of the 1600 KC section but outside the
passband of the low IF from overloading the 1600KC section.
b) aim for about -40dBm out of the detector for the minimum discernable signal.
This is about 2.2 mV RMS into 50 ohms or 31 mV into 10K.
c) The above gives about 40 to 50 dB of gain needed in the IF to be able to
receive a -100dBm signal at the antenna terminals. If I remember correctly this
is pretty close to the limit for low HF (40meters).
I hope this has been useful.
Ray Mack
WD5IFS
mack@mails.imed.com
Friendswood (Houston), TX
______________________________ Reply Separator _________________________________
<snip>
1)
Circuit starts with a 6BJ6 Pentode as RF amp. Plug in coils for each
band. Antenna link coupled to grid coil which is tuned by one of three
sections of the 5-23mmf/section VFO cap. RF gain via cathode bias pot.
Mosley manual said triode amp is better on high bands owing to
noise considerations. Real feature or just making what they
used look good?
<snip>
2)
How to determine best VFO injection level?
<snip>
3)
Here I'm curious as to how to determine whether ot not the
amount of coupling I have is right, too much, or not enough.
Would a 1600kc xtal (there is one in the junk box) be too
tight as a filter or attenuate the signal too much at this stage
of the RX?
<snip>
4)
Deja-vu, the second IF amp which is followed by the third of the triplet
of transformers. The newly shaped IF signal is then routed to either a
6BE6 product detector or 1/2 of a 12AU7 double triode as envelope
detector. Both IF gains are via one cathode bias pot.
As an operator, what is a good way to tell how much IF gain
is enough.
<snip>
- -bob
Date: Wed, 4 Jun 1997 17:21:24 +0000
From: "Brian Carling, Radio AF4K" <bry@mnsinc.com>
Subject: Mic. connectors
In abind for one of those "BOATANCHOR" type screw-on microphone
plugs?
You kow, the kind that Heath and others used to us, with a center pin
that is littl emore than a "nub" about 3/32" across, and a screw on
shield ring that is about 1/2" threads?
Well, this is amazing but I read in a n old Hints and Kinks, that you
can use a PL-259 plug for on e of those things in a pinch!
Now some of these were spring loaded plugs, others not,
but all used the same size threads and pin scheme with one single
center conductor pin, usually in some sort of bakelite slug.
They say you just cut off most of the center pin and it will screw
right on to the mic. connector!
I haven't tried thise but it sounded like a good wheeze!
On my DX-60 the previous owner took out that panel jack and put in a
more "modern" 4 pin jack of the type used on the 1970s
era rice boxes like the FT-101 series etc.
Any comments? Bry, AF4K
****************************************************
*** 73 from Radio AF4K/G3XLQ Gaithersburg, MD USA *
** E-mail to: bry@mnsinc.com *
*** See the interesting ham radio resources at: *
** http://www.mnsinc.com/bry/ *
****************************************************
Date: Wed, 4 Jun 1997 17:21:24 +0000
From: "Brian Carling, Radio AF4K" <bry@mnsinc.com>
Subject: Re: Computer Power Supplies
On 4 Jun 97 at 9:31, Dexter Francis spoke about Computer Power
Supplies and said:
> Many computer supplies are switch mode regulators that double or
> triple the line voltage before running it thru the
> chopper/modulatior section. This means that there is probably a
> place to get 240 to 360 volts out of these little fellows, in
> addition to the +12 and +5. A little fiddling with the +5 regulator
> could probably make it yield +6 and voila! an integrated HV and LV
> power supply for your next gassless valve project.
Yes, IF you can stand the QRM from that darn switching mode supply
that will likely appear in your receiver!
Any time a computer is on the same room with most sensitive HF
receivers I have tried you will often get a BUZZ every 50 kHz or so
up and down all of the HF bands!!
****************************************************
*** 73 from Radio AF4K/G3XLQ Gaithersburg, MD USA *
** E-mail to: bry@mnsinc.com *
*** See the interesting ham radio resources at: *
** http://www.mnsinc.com/bry/ *
****************************************************
Date: Wed, 4 Jun 1997 18:33:16 -0400 (EDT)
From: "Walter L. Marshall" <wmarshal@CapAccess.org>
Subject: Re: Mic. connectors
That's what I do. Works fine.
Walter
Date: Wed, 4 Jun 1997 18:34:22 -0400 (EDT)
From: "Walter L. Marshall" <wmarshal@CapAccess.org>
Subject: Re: Mic. connectors
I also file the end smooth.
Walter
Date: Wed, 4 Jun 1997 18:59:48 -0500 (CDT)
From: Bob Roehrig <broehrig@admin.aurora.edu>
Subject: Re: Computer Power Supplies
On Wed, 4 Jun 1997, Brian Carling, Radio AF4K wrote:
>
> Any time a computer is on the same room with most sensitive HF
> receivers I have tried you will often get a BUZZ every 50 kHz or so
> up and down all of the HF bands!!
A good deal of the computer generated RF is from the monitor. Stuff can
also leak out the serial ports and any other connector cable attached to
the PC. Most printers also have switching supplies and are usually plastic
cased and as a result can radiate much worse then the PC itself.
A lot of the PC generated garbage can be reduced by wrapping the attached
cords thru toroid cores.
E-mail broehrig@admin.aurora.edu 73 de Bob, K9EUI
CIS: Data / Telecom Aurora University, Aurora, IL
630-844-4898 Fax 630-844-5530
Date: Wed, 4 Jun 1997 17:16:37 -0700 (PDT)
From: Ken Gordon <keng@uidaho.edu>
Subject: FT-243 to HC-6
Oops! Got it the wrong way around. Sorry. Many years ago, I had an
adapter made out of a small piece of rectangular lucite. It had two of
the small HC-6 size pins mounted in it and two FT-243 size holes drilled
along side of the HC-6 pins. Just plug the FT-243 into it and then plug it
into the transmitter. Shouldn't be too hard to make one if you have
access to a drill press, a couple of nails the same diameter at the HC-6
pins, and a bit of lucite.
Ken W7EKB
Date: Wed, 4 Jun 1997 21:32:19 -0500 (CDT)
From: wbfim@echo.sound.net
Subject: MANUAL NEEDED
I just received a regen set in very good condition but I in dire need of a
manual for it. I hope some one can help.
It is a Federal telephone and Radio Corp. Model No RC 123. It tunes from 15
kcs to 650 kcs. The contract says that it was made for the Coast Guard.
Thanks for your time
Dennis
KE0QM
Date: Wed, 4 Jun 1997 21:32:23 -0500 (CDT)
From: wbfim@echo.sound.net
Subject: SCRAP BOOK BY W2IMB
I am looking to make a collection of 'The low and meduim frequency radio
scrap book' by W2IMB.
I would take origionals or copies. I have the 3rd, 7th, and 9th editions.
Thanks and 73/72
Dennis
KE0QM
Date: Wed, 4 Jun 1997 22:35:40 -0600 (MDT)
From: toyboat@freenet.edmonton.ab.ca
Subject: Re: Lorenz Type Coils
On Wed, 4 Jun 1997 EWoodman@aol.com wrote:
> Shane,
> I saw you mention that you have used some Lorenz-type coils in your
> transmitter. I've always liked the looks of those. Do the standard coil
> winding formulas work for this type of coil?
>
> 73 Eric KA1YRV
>
An excellent question! Unfortunately, I have not the foggiest idea.
Considering the weird woven geometric pattern of them, I doubt that
any simple solenoid coil formula would adapt. I simply used the
published project data.
I do know that they have excellent Q, are very easy to wind, and work
just fine. They slide off the dowels easy, yet can be positioned
easily and firmly. I glued my dowels into the wooden chassis and
made a removable tensioner top cap of a 3" X 1/2" circle of poplar,
drilled with matching holes for the dowels. I used waxed linen
leather stitching thread (Tandy) to tie the coils.
Perhaps someone else can come up with the formulas for these?
I'd like to know them myself.
Regards,
|\
| \
| \
| \
|____\
_______|______\O/_
\ /
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Shane <toyboat@freenet.edmonton.ab.ca>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
End of glowbugs V1 #51
**********************
Created by Steve Modena, AB4EL
Comments and suggestions to modena@SunSITE.unc.edu