开云体育

Hi all,

Here's a tale of a diabolical fault I recently encountered in a piece
of classic Tek equipment. The experienced Tek people here might
say "I knew about that" but it was a great achievement for me!

The patient was my youngest 1A1 plugin, with FET inputs and lever
input switches. The symptoms - low gain on both channels on switching
on, with further gradual drop in gain as the unit warmed up over 5-10
minutes. DC balance was preserved.

For those of you not familiar with the 1A1 (what are you doing here?)
it's a standard 2 channel vertical amplifier for the 530 to 550
series. Early models had nuvistor inputs and later models have FET
inputs. Each channel has an amplifier board with the transistor
amplifiers following the input stage and there is a common output
card with the chop/alt/add circuitry and a hybrid cascode output
using nuvistors for the final grounded grid stages. Amplification is
differential throughout.

Since the fault was common to both channels I initially thought it
was in the output card and spent a lot of time swapping my small
stock of nuvistors and probing around, all to no avail. The clue to
the solution came when I substituted boards from a known good 1A1. I
found that swapping the channel 2 board cured the problem. Further
measurements with the original card showed that the 11 volt supply
was high at 12.25 volts. Because the fault seemed temperature
sensitive I tried using freeze spray and found that cooling
transistors Q253A&B (a pair of 2N3563s) temporarily restored
performance on channel 2. The base bias for these 2 was derived from
the +11 volt rail and was too high, causing excess collector current.

Where does the +11 volts come from? Tek designs of this era were
notorious for clever and tricky use of the power supplies. Plugins
are provided with the standard scope supplies at the connector and
one pin has to sink 150 mA at 75 volts. This is derived from the +100
volt supply in the scope dropped through a couple of 12.6 volt tube
heaters. In different plugins it is used for a series heater string
and miscellaneous DC supplies. My 1A1 has 2 tubes and this accounts
for 12.6 volts of the 75. The rest is dropped through a string of
resistors and provides +39, +11 and +5 volt rails. The 11 volt rail
supplies grid bias for the output tubes and is also the collector
supply for the transistors feeding Q253A&B from where it affects the
bias on all the following amplifiers.

Tracing through the divider, I found R496 and R497, both 130 ohms 5%
half watt and in series, had drifted high. Not by much, they were 140
and 147 ohms when cold but this was enough to increase the bias which
caused thermal runaway of Q253A&B. I checked this by temporarily
shunting them with 2.7K which I calculated would reduce the total
resistance to the proper value - this fixed the problem instantly! I
replaced them permanently with 1% half watt metal film resistors, the
11 volt supply is now at 11.4 volts and the plugin works perfectly.

Why did swapping the Ch 2 card cure the problem? The two resistors
are each on one of the input cards. Swapping one card reduced the
total resistance just enough to stop the problem and point me to what
was going on. Why was DC balance preserved? Because the fault was
common mode, and fortunately didn't upset the operating conditions of
the output tubes.

What a sneaky fault! Who would have imagined a problem in the output
tube heater supply would cause thermal runaway of a couple of pairs
of transistors a few stages back which would in turn upset
differential gain but not DC balance? It shows how carefully you have
to understand how this complex circuitry works. It also shows how
interlinked all the various parts of the circuit are - there's no
such thing as an isolated stage for troubleshooting purposes.

Since I was using a 556 (850 watts consumption) for simultaneously
supplying the 1A1 and troubleshooting it, it was a hot experience so
a cold beer was very welcome afterwards!!

Cheers,

Morris

Morris

That's a keeper.

Jim Nunn

jimnunn0000@...

I'll "second" that Jim. That one is worth printing (already done).

Nice work Morris and thanks for the details and explanation. I learned a lot from that post.

Phil

That's a new one on me, Morris. Congratulations and thanks for sharing that
info.

Stan
w7ni@...

morriso2002 wrote:

Hi,

that story reminds me a bit about onother story were 2 identical parts in
separate circuits
failed at the same time in the same way. In my collection I have also a
Philips PM3230, a real
dual beam scope with tubes. After the scope hasn't been used for a while, I
wasn't able to
turn the brighness down. And that was the case for both beams. I spent hours
in
measuring in the HV supply circuits common for both electron guns, but
everything was OK.
Then I started looking at the voltages of one individual gun. What I found:
The Z-Axis modulation is done via the grid. A capacitor is connected between
the grid and
the Z-Axis input on the back. On the Z-Axis connector side of the capacitor
a 1 MOhm resistor
is shunted to GND. That's a common construction used in many scopes. When I
removed
the cap, the brightness control worked for that particular channel. I
measured across the
cap and found it had a resistance of several 100 Kohms. And that was exactly
also the case
in the 2nd channel. Both parts failed at the same time in the same way. The
leaking cap in series
with the 1MOhm resisitor was pulling up the grid voltage towards GND and was
overriding the
intensity control

Leo




-----Ursprüngliche Nachricht-----
Von: morriso2002 [mailto:morriso2002@...]
Gesendet: Donnerstag, 27. Dezember 2001 23:56
An: TekScopes@...
Betreff: [TekScopes] Fun with the 1A1


Hi all,

Here's a tale of a diabolical fault I recently encountered in a piece
of classic Tek equipment. The experienced Tek people here might
say "I knew about that" but it was a great achievement for me!

The patient was my youngest 1A1 plugin, with FET inputs and lever
input switches. The symptoms - low gain on both channels on switching
on, with further gradual drop in gain as the unit warmed up over 5-10
minutes. DC balance was preserved.

For those of you not familiar with the 1A1 (what are you doing here?)
it's a standard 2 channel vertical amplifier for the 530 to 550
series. Early models had nuvistor inputs and later models have FET
inputs. Each channel has an amplifier board with the transistor
amplifiers following the input stage and there is a common output
card with the chop/alt/add circuitry and a hybrid cascode output
using nuvistors for the final grounded grid stages. Amplification is
differential throughout.

Since the fault was common to both channels I initially thought it
was in the output card and spent a lot of time swapping my small
stock of nuvistors and probing around, all to no avail. The clue to
the solution came when I substituted boards from a known good 1A1. I
found that swapping the channel 2 board cured the problem. Further
measurements with the original card showed that the 11 volt supply
was high at 12.25 volts. Because the fault seemed temperature
sensitive I tried using freeze spray and found that cooling
transistors Q253A&B (a pair of 2N3563s) temporarily restored
performance on channel 2. The base bias for these 2 was derived from
the +11 volt rail and was too high, causing excess collector current.

Where does the +11 volts come from? Tek designs of this era were
notorious for clever and tricky use of the power supplies. Plugins
are provided with the standard scope supplies at the connector and
one pin has to sink 150 mA at 75 volts. This is derived from the +100
volt supply in the scope dropped through a couple of 12.6 volt tube
heaters. In different plugins it is used for a series heater string
and miscellaneous DC supplies. My 1A1 has 2 tubes and this accounts
for 12.6 volts of the 75. The rest is dropped through a string of
resistors and provides +39, +11 and +5 volt rails. The 11 volt rail
supplies grid bias for the output tubes and is also the collector
supply for the transistors feeding Q253A&B from where it affects the
bias on all the following amplifiers.

Tracing through the divider, I found R496 and R497, both 130 ohms 5%
half watt and in series, had drifted high. Not by much, they were 140
and 147 ohms when cold but this was enough to increase the bias which
caused thermal runaway of Q253A&B. I checked this by temporarily
shunting them with 2.7K which I calculated would reduce the total
resistance to the proper value - this fixed the problem instantly! I
replaced them permanently with 1% half watt metal film resistors, the
11 volt supply is now at 11.4 volts and the plugin works perfectly.

Why did swapping the Ch 2 card cure the problem? The two resistors
are each on one of the input cards. Swapping one card reduced the
total resistance just enough to stop the problem and point me to what
was going on. Why was DC balance preserved? Because the fault was
common mode, and fortunately didn't upset the operating conditions of
the output tubes.

What a sneaky fault! Who would have imagined a problem in the output
tube heater supply would cause thermal runaway of a couple of pairs
of transistors a few stages back which would in turn upset
differential gain but not DC balance? It shows how carefully you have
to understand how this complex circuitry works. It also shows how
interlinked all the various parts of the circuit are - there's no
such thing as an isolated stage for troubleshooting purposes.

Since I was using a 556 (850 watts consumption) for simultaneously
supplying the 1A1 and troubleshooting it, it was a hot experience so
a cold beer was very welcome afterwards!!

Cheers,

Morris


Yahoo! Groups Sponsor
ADVERTISEMENT




To unsubscribe from this group, send an email to:
TekScopes-unsubscribe@...



Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

Hello Morris,

That was quite a thing that you found, really something to be proud of.

I am curious what was voltage drop across each of these two 130 Ohm resistors. I have seen a number of resistors in Tek instruments that were discolored from overheating, because they were undersized. In an 1A1, the resistors would be probably carbon composition, a tough beast to kill, but they do not show discoloration either, so you would not have a visual indication that there is something wrong.

If those two resistors were really overloaded, then circuitry graduates from complex to a kluge. Please, let me know what was voltage drop.

Regards

Miroslav Pokorni

morriso2002 <morriso2002@...> wrote: Hi all,

Here's a tale of a diabolical fault I recently encountered in a piece
of classic Tek equipment. The experienced Tek people here might
say "I knew about that" but it was a great achievement for me!

The patient was my youngest 1A1 plugin, with FET inputs and lever
input switches. The symptoms - low gain on both channels on switching
on, with further gradual drop in gain as the unit warmed up over 5-10
minutes. DC balance was preserved.

For those of you not familiar with the 1A1 (what are you doing here?)
it's a standard 2 channel vertical amplifier for the 530 to 550
series. Early models had nuvistor inputs and later models have FET
inputs. Each channel has an amplifier board with the transistor
amplifiers following the input stage and there is a common output
card with the chop/alt/add circuitry and a hybrid cascode output
using nuvistors for the final grounded grid stages. Amplification is
differential throughout.

Since the fault was common to both channels I initially thought it
was in the output card and spent a lot of time swapping my small
stock of nuvistors and probing around, all to no avail. The clue to
the solution came when I substituted boards from a known good 1A1. I
found that swapping the channel 2 board cured the problem. Further
measurements with the original card showed that the 11 volt supply
was high at 12.25 volts. Because the fault seemed temperature
sensitive I tried using freeze spray and found that cooling
transistors Q253A&B (a pair of 2N3563s) temporarily restored
performance on channel 2. The base bias for these 2 was derived from
the +11 volt rail and was too high, causing excess collector current.

Where does the +11 volts come from? Tek designs of this era were
notorious for clever and tricky use of the power supplies. Plugins
are provided with the standard scope supplies at the connector and
one pin has to sink 150 mA at 75 volts. This is derived from the +100
volt supply in the scope dropped through a couple of 12.6 volt tube
heaters. In different plugins it is used for a series heater string
and miscellaneous DC supplies. My 1A1 has 2 tubes and this accounts
for 12.6 volts of the 75. The rest is dropped through a string of
resistors and provides +39, +11 and +5 volt rails. The 11 volt rail
supplies grid bias for the output tubes and is also the collector
supply for the transistors feeding Q253A&B from where it affects the
bias on all the following amplifiers.

Tracing through the divider, I found R496 and R497, both 130 ohms 5%
half watt and in series, had drifted high. Not by much, they were 140
and 147 ohms when cold but this was enough to increase the bias which
caused thermal runaway of Q253A&B. I checked this by temporarily
shunting them with 2.7K which I calculated would reduce the total
resistance to the proper value - this fixed the problem instantly! I
replaced them permanently with 1% half watt metal film resistors, the
11 volt supply is now at 11.4 volts and the plugin works perfectly.

Why did swapping the Ch 2 card cure the problem? The two resistors
are each on one of the input cards. Swapping one card reduced the
total resistance just enough to stop the problem and point me to what
was going on. Why was DC balance preserved? Because the fault was
common mode, and fortunately didn't upset the operating conditions of
the output tubes.

What a sneaky fault! Who would have imagined a problem in the output
tube heater supply would cause thermal runaway of a couple of pairs
of transistors a few stages back which would in turn upset
differential gain but not DC balance? It shows how carefully you have
to understand how this complex circuitry works. It also shows how
interlinked all the various parts of the circuit are - there's no
such thing as an isolated stage for troubleshooting purposes.

Since I was using a 556 (850 watts consumption) for simultaneously
supplying the 1A1 and troubleshooting it, it was a hot experience so
a cold beer was very welcome afterwards!!

Cheers,

Morris


Yahoo! Groups Sponsor
To unsubscribe from this group, send an email to:
TekScopes-unsubscribe@...



Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.



---------------------------------
Do You Yahoo!?
Send your FREE holiday greetings online at Yahoo! Greetings.

Hi Miroslav,

Thanks for the message

--- In TekScopes@y..., Miroslav Pokorni <mpokorni2000@y...> wrote:

That was quite a thing that you found, really something to be proud

of.

I am curious what was voltage drop across each of these two 130 Ohm

resistors.

If those two resistors were really overloaded, then circuitry

graduates from complex to a kluge. Please, let me know what was
voltage drop.

It was no kludge, just a faulty batch of 130 ohm resistors I suspect.
They are in series from the 11 volt supply to ground so each had 5.5
volts across them. That's less than 0.25 watts.

Morris