[R-390] Ballatube regulator simulations
Drew Papanek
[email protected]
Sun, 29 Dec 2002 18:15:46 -0500
Hello All,
I ran a computer simulation of a couple of different BallasTube replacement
circuits. These are of the recently discussed "clipped sinewave" variety.
No filter capacitors are used (beyond the small caps recommended for LM317
stability). The LM317 in each case sees raw unfiltered DC straight from the
dead horse's, er, rectifier's mouth.
They are not true RMS current regulators but for some situations might
provide
a reasonable approximation. I found the results to be interesting and maybe
you will find them to be boring.
Circuit Descriptions:
Regulator circuit #1: This circuit was designed, built, and tested by Dr.
Gerald
Johnson and reported on the R-390A list a while back. It consists of full
wave
bridge rectifier wrapped around LM317 configured as DC current regulator.
Current sense resistor for LM317 may be preset for I=347 mA using a quiet DC
source of about 6 volts at AC terminals of bridge rectifier. Rectifier
AC terminals are then disconnected from DC source and connected to
BallasTube
socket pins 2 and 7. This should yield RMS heater current as specified at
"Line
Nominal" conditions. Because LM317's reference voltage may lie between 1.2
and
1.3 volts, current sense resistor value can range from 3.46 to 3.75 ohms.
Tube heaters see a clipped AC sine wave.
Regulator circuit #2: This circuit was recently proposed by Jim Shorney and
as
far as I know has not been tested. Circuit consists of half wave rectifier
feeding LM317 configured as DC current regulator. Current sense resistor
for
LM317 may be preset for I=523 mA using DC source as for circuit #1 above.
Rectifier is fed from ballasTube socket pin 2 (Vsec) and regulated DC output
fed to tube heaters at BallasTube socket pin 7. Current sense resistor for
LM317 can range from 2.29 to 2.49 ohms. Tube heaters see pulsating half
wave
rectified DC with clipped peaks.
Simulation Conditions:
1.Ihtr is RMS current through seriesed 6BA6 PTO and BFO tube heaters. Total
resistance is assumed to be 42 ohms hot.
2.Vsec is RMS AC voltage from transformer secondary winding at BallasTube
pin 2.
3.Dev is Ihtr deviation in percent from Ihtr specified at Line Nominal
conditions.
4.Pd reg is LM317 regulator power dissipation in watts.
5.For Startup, line voltage and Vsec are as specified for Line Nominal
conditions and initial heater total resistance is assumed to be 5.4 ohms
cold.
Simulation results for circuit #1:
Line Nominal:AC line=115, Vsec=25.2, Ihtr=300mA (adjusted), Pd reg=2.9w
Line-8.7%: AC line=105, Vsec=23.0, Ihtr=295mA, dev= -1.7%, Pd reg=2.3w
Line+8.7%: AC line=125, Vsec=27.4, Ihtr=304mA, dev= +1.3%, Pd reg=3.5w
Line-13%: AC line=100, Vsec=21.9, Ihtr=292mA, dev= -2.6%, Pd reg=2.0w
Line+13%: AC line=130, Vsec=28.5, Ihtr=306mA, dev= +2.0%, Pd reg=3.8w
Startup: Ihtr=330 mA
Simulation results for circuit #2:
Line Nominal:AC line=115VAC, Vsec=25.2, Ihtr=300mA (adjusted), Pd reg=1.2w
Line-8.7%: AC line=105, Vsec=23.0, Ihtr=291mA, dev= -2.9%, Pd reg=.8w
Line+8.7%: AC line=125, Vsec=27.4, Ihtr=307mA, dev= +2.3%, Pd reg=1.6w
Line-13%: AC line=100, Vsec=21.9, Ihtr=286mA, dev= -4.7%, Pd reg=.7w
Line+13%: AC line=130, Vsec=28.5, Ihtr=310mA, dev= +3.3%, Pd reg=1.8w
Startup: Ihtr=352 mA
Notes and Musings:
1. Precision regulators these ain't! The 3TF7 BallasTube does better with
its
specification of +-1% over a much wider line voltage range than shown in
these
simulations. An LM317 operated with sufficient headroom from a filtered DC
source does far better still. The sophisticated 3DW7 Tubester with its
microprocessor control is reputed to be excellent. However, circuits #1 and
#2 do considerably better than no regulator at all.
2. Some like it hot! In order of least to most heat generation would be the
cool running 3DW7, then circuit #2, 3TF7, circuit #1, filtered DC operated
LM317. From a standpoint of simple regulator heatsinking, circuit #2 might
be
attractive.
3. It has been argued that current regulation for the PTO and BFO tube
heaters
is not really necessary for the type of operation that most of our R-39x
see.
There are several schemes which eliminate the 3TF7 current regulator and
supply
unregulated PTO and BFO heater power. Proponents of these schemes claim
that
frequency stability is still excellent. More data on frequency variation vs
PTO/BFO heater current variation would help to resolve doubt in a given
usage
situation. If one were to classify stability with unregulated heater power
as
"quite good", perhaps circuit #1 or #2 could upgrade the classification to
"very good" or "excellent". Schemes offering still better regulation might
only
provide improvements past the point of diminishing returns. For some users,
however, anything less than the best obtainable would be unacceptable.
4.If you want original performance, use an original part! Put those 3TF7's
to
work.
5.Note that cold startup current is a bit higher than current when heaters
have
warmed up. Compare the values above to the 2.3 amps a cold 6BA6 would draw
from
a supply without current regulation. Current limiting action reduces heater
inrush current and resultant heater stress. This might help prolong tube
life.
6. Posting of these results helps prove that even a "dead horse" can still
make
a pile of manure! As such, these ramblings should be taken as just that: a
product of my "fertile" imagination. Use these ideas at your own risk. My
liability is limited to the saying of a requiem for deceased 6BA6's.
7.For my R-390A, I prefer and use the "Two 12BA6's and a paperclip" method
for
its utter simplicity and because I'm lazy. Maybe someday I'll test circuits
#1
and #2 to satisfy my curiosity.
Neither 6BA6's nor electrolytic filter caps were harmed during the running
of
these simulations.
Drew
_________________________________________________________________
The new MSN 8: smart spam protection and 3 months FREE*.
http://join.msn.com/?page=features/junkmail&xAPID=42&PS=47575&PI=7324&DI=7474&SU=
http://www.hotmail.msn.com/cgi-bin/getmsg&HL=1216hotmailtaglines_smartspamprotection_3mf