"MoiInAust" wrote in message
...

I have an early AR88 that developed a very elusive fault
while I was coincidentally changing some capacitors and
resistors. Of course I suspected I may have miswired
something but I have checked and rechecked and it doesn't
seem so. May be just a coincidence.

This fault has defied many weeks of painstaking work. I
wonder if members of the list have any ideas

On the excellent test table provided in EMER 773 (on the
VMARS website), under the specified test conditions
(including AVC off) there should be voltages of - 1.2 V on
the grids of V1, V2, V5, V6 and the anode (plate) and
cathode of V8b, taken to the slider of RV3 (RF gain) when
at max. This is of course the AVC line, and for the
voltage as shown to the slider of RV3 there must be 1.2v
across resistor R42 (390K). Well those readings used to be
obtained, but after my work they are very slightly
positive! (I emphasise that the readings are not to
chassis but to the slider of RV3 as instructed. looking at
the circuit diagram, to get that voltage between those
points, there must be a very small current flowing through
R42. It would have to be as a result of voltage from the
bias dropper chain of R45, R44, and R43. These have all
been checked OK and there is approx -30 volts to chassis
at the end of the chain (total HT/B current for the set
110mA). With v8b shorted (per EMER 773) the other end of
R42 goes via R47 (2.2 Meg) RV1 (66K) and R39 (33K). All
those have been checked OK but no volts across R42!

I have tried disconnecting the AVC main feed to V1 etc
from SW22, but no difference so the problem is not in any
of the circuitry before v8b.

Next problem (connected?) There is mild AVC action
*whether the AVC switch is on or off*!

Any fresh ideas?

BTW, to get -1.2 volts across R42 (390 K) would require a
very small current of .003 mA. How is that obtained if
there's a chain through R42, R47, RV1, R49?

Wouldn't strain your mind with this, but you did ask!
Cheers

I'll follow along the schematic and see if I have any
ideas. RCA's schematics can be confusing:-(



--
Richard Knoppow
Los Angeles
WB6KBL

"Richard Knoppow" wrote in message
m...

"MoiInAust" wrote in message
...

I have an early AR88 that developed a very elusive fault while I was
coincidentally changing some capacitors and resistors. Of course I
suspected I may have miswired something but I have checked and rechecked
and it doesn't seem so. May be just a coincidence.

This fault has defied many weeks of painstaking work. I wonder if members
of the list have any ideas

On the excellent test table provided in EMER 773 (on the VMARS website),
under the specified test conditions (including AVC off) there should be
voltages of - 1.2 V on the grids of V1, V2, V5, V6 and the anode (plate)
and cathode of V8b, taken to the slider of RV3 (RF gain) when at max.
This is of course the AVC line, and for the voltage as shown to the
slider of RV3 there must be 1.2v across resistor R42 (390K). Well those
readings used to be obtained, but after my work they are very slightly
positive! (I emphasise that the readings are not to chassis but to the
slider of RV3 as instructed. looking at the circuit diagram, to get that
voltage between those points, there must be a very small current flowing
through R42. It would have to be as a result of voltage from the bias
dropper chain of R45, R44, and R43. These have all been checked OK and
there is approx -30 volts to chassis at the end of the chain (total HT/B
current for the set 110mA). With v8b shorted (per EMER 773) the other end
of R42 goes via R47 (2.2 Meg) RV1 (66K) and R39 (33K). All those have
been checked OK but no volts across R42!

I have tried disconnecting the AVC main feed to V1 etc from SW22, but no
difference so the problem is not in any of the circuitry before v8b.

Next problem (connected?) There is mild AVC action *whether the AVC
switch is on or off*!

Any fresh ideas?

BTW, to get -1.2 volts across R42 (390 K) would require a very small
current of .003 mA. How is that obtained if there's a chain through R42,
R47, RV1, R49?

Wouldn't strain your mind with this, but you did ask! Cheers

I'll follow along the schematic and see if I have any ideas. RCA's
schematics can be confusing:-(

Well I agree. Being a Pom, I used the Anlglicized version of the circuit
which is on EMER772. I find the convention of having high voltage lines
higher than low ones (and the negative AVC line below the chassis 0) easier
to understand. The US diagram in the original handbook has the valves upside
down to my mind (!) Having said that, I consider the AR88 (US design that
it is) to have been decades ahead of any Pom effort (like the infamous 38
set!). 1940 or even earlier and it's still a beaut performer.

"MoiInAust" wrote in message
...

"Richard Knoppow" wrote in
message
m...

"MoiInAust" wrote in message
...

I have an early AR88 that developed a very elusive fault
while I was coincidentally changing some capacitors and
resistors. Of course I suspected I may have miswired
something but I have checked and rechecked and it
doesn't seem so. May be just a coincidence.

This fault has defied many weeks of painstaking work. I
wonder if members of the list have any ideas

On the excellent test table provided in EMER 773 (on the
VMARS website), under the specified test conditions
(including AVC off) there should be voltages of - 1.2 V
on the grids of V1, V2, V5, V6 and the anode (plate) and
cathode of V8b, taken to the slider of RV3 (RF gain)
when at max. This is of course the AVC line, and for the
voltage as shown to the slider of RV3 there must be 1.2v
across resistor R42 (390K). Well those readings used to
be obtained, but after my work they are very slightly
positive! (I emphasise that the readings are not to
chassis but to the slider of RV3 as instructed. looking
at the circuit diagram, to get that voltage between
those points, there must be a very small current flowing
through R42. It would have to be as a result of voltage
from the bias dropper chain of R45, R44, and R43. These
have all been checked OK and there is approx -30 volts
to chassis at the end of the chain (total HT/B current
for the set 110mA). With v8b shorted (per EMER 773) the
other end of R42 goes via R47 (2.2 Meg) RV1 (66K) and
R39 (33K). All those have been checked OK but no volts
across R42!

I have tried disconnecting the AVC main feed to V1 etc
from SW22, but no difference so the problem is not in
any of the circuitry before v8b.

Next problem (connected?) There is mild AVC action
*whether the AVC switch is on or off*!

Any fresh ideas?

BTW, to get -1.2 volts across R42 (390 K) would require
a very small current of .003 mA. How is that obtained
if there's a chain through R42, R47, RV1, R49?

Wouldn't strain your mind with this, but you did ask!
Cheers

I'll follow along the schematic and see if I have any
ideas. RCA's schematics can be confusing:-(

Well I agree. Being a Pom, I used the Anlglicized version
of the circuit which is on EMER772. I find the convention
of having high voltage lines higher than low ones (and the
negative AVC line below the chassis 0) easier to
understand. The US diagram in the original handbook has
the valves upside down to my mind (!) Having said that, I
consider the AR88 (US design that it is) to have been
decades ahead of any Pom effort (like the infamous 38
set!). 1940 or even earlier and it's still a beaut
performer.
The RCA schematics are sort of combination schematic
and wiring diagram with all the pins shown in actual order.
A lot of schematics for home radios were drawn like this.
Its confusing because there is no natural flow. My other
favorite is Western Electric transmitter diagrams with all
the filaments and ground wires shown so they look like a map
of a large freight yard.

BTW, I downloaded a bunch of stuff from the VMARS site,
lots of R-390 and other books there and a military handbook
on the older Super-Pro models.

--
Richard Knoppow
Los Angeles
WB6KBL

"Richard Knoppow" wrote
BTW, I downloaded a bunch of stuff from the VMARS site, lots of R-390
and other books there and a military handbook on the older Super-Pro
models.

Good site, isn't it! Now Richard, stop salivating over the R-390 stuff and
solve my problem with the AR88!

"MoiInAust" wrote in message
...

"Richard Knoppow" wrote
BTW, I downloaded a bunch of stuff from the VMARS site,
lots of R-390 and other books there and a military
handbook on the older Super-Pro models.

Good site, isn't it! Now Richard, stop salivating over
the R-390 stuff and solve my problem with the AR88!
I redrew the circuit. The AVC circuit is fairly
conventional. The AVC voltage comes from the detector and
the second half of the 6H6 is used to generate a "delay"
voltage to prevent the AVC from acting until the signal
strength reaches a certain minimum level. This is done to
maintain the RF and IF stages at full gain. There is a
negative bias supply obtained from a series of resistors in
the center-tap return of the rectifier. The RF gain voltage
comes from the RF gain control which is in this string.
About the only way I can see for a positive voltage to
appear on the AVC-RF gain bus is if this whole string is
being pulled positive. If that's so it should also show up
on the audio tubes which are biased from this source. Unless
you made a gross error in wiring I would be suspicious of
C-96 and C97, the first two filter caps since they are
returned to the top of the negative supply rather than to
chassis ground. C-98, the last cap, goes to chassis ground
so can't cause trouble. If one or both of the the filter
caps is leaky it could be pulling the negative supply to a
positive value.
I wouldn't worry about gassy tubes. Positive voltage
can come from one of the controlled tubes becoming gassy but
they are pretty well isolated in this circuit. I can't
account for the slight AVC action with the AVC switch in
manual but if the negative supply has been pulled positive
it might affect this.
The AVC switch essentially shorts out the AVC and puts
the controlled stages directly on the negative supply.
However, the whole AVC bus floats on the negative supply so
that the RF gain control works in both AVC and Manual.

I certainly agree with those who suggest going over
your work carefully. If you know another knowlegible person
have them look also since its possible for someone to
overlook the same thing over and over where someone else
will spot it right away. But, check the filter caps first.



--
Richard Knoppow
Los Angeles
WB6KBL