HV Multiplier defective in 7603?


raymonddompfrank <r.domp.frank@...>
 

Hi all,
My recently restored and very nice early Heerenveen (NL-) made 7603 has developed a problem:
After working OK for a few days, suddenly brightness and focus deteriorated, came back again, deteriorated again and finally the screen became completely dark, all this within 5 minutes. After I opened the case, it immediately became apparent that the HV cage was quite hot, I guess around 50C. Most likely, that's because of Q1216 and Q1218 heating up. No burning smell.

Looking for the fault, being careful and full of respect for HV, I did the following:

1. Disconnected HV cage and measured all LV voltages (-50V, -15V, +5V, +15V, +5-V, +130V), using a 'scope with DVM). All ok, ripple very low.

2. Reconnected HV cage. LV's remain ok, HV Test point (should be -2975V)reaches ca -1.8KV, next goes further negative, something like -2.5KV, then back to around -1.8KV etc. with a period of 1 or 2 seconds. Sort of tick mode below -1.8KV. LV's remain OK, so the 'scope isn't in tick mode. PDA voltage sort of follows change of HV (used P6015). No brightness.

3. Unplugged and discharged post deflection anode. 'Scope behaved as in 2.

4. Disconnected HV Multiplier (U1230) at its input. HV test point reaches almost -4KV and remains steady.

To me, this looks like a HV breakdown in the multiplier.
It seems that these multipliers are rather expensive at Qservice etc. so I'm considering buying one second hand locally or building one myself. Apart from obvious things like HV isolation (air), >10KV C's and >7KV(?) diodes, would there be any special things to consider when building a multiplier myself?

Any thoughts at all on this?

Thanks!

Raymond


raymonddompfrank <r.domp.frank@...>
 

Under 3.: I *did* switch on again here, now without PDA connected...

--- In TekScopes@yahoogroups.com, "raymonddompfrank" <r.domp.frank@...> wrote:

Hi all,
My recently restored and very nice early Heerenveen (NL-) made 7603 has developed a problem:
After working OK for a few days, suddenly brightness and focus deteriorated, came back again, deteriorated again and finally the screen became completely dark, all this within 5 minutes. After I opened the case, it immediately became apparent that the HV cage was quite hot, I guess around 50C. Most likely, that's because of Q1216 and Q1218 heating up. No burning smell.

Looking for the fault, being careful and full of respect for HV, I did the following:

1. Disconnected HV cage and measured all LV voltages (-50V, -15V, +5V, +15V, +5-V, +130V), using a 'scope with DVM). All ok, ripple very low.

2. Reconnected HV cage. LV's remain ok, HV Test point (should be -2975V)reaches ca -1.8KV, next goes further negative, something like -2.5KV, then back to around -1.8KV etc. with a period of 1 or 2 seconds. Sort of tick mode below -1.8KV. LV's remain OK, so the 'scope isn't in tick mode. PDA voltage sort of follows change of HV (used P6015). No brightness.

3. Unplugged and discharged post deflection anode. 'Scope behaved as in 2.

4. Disconnected HV Multiplier (U1230) at its input. HV test point reaches almost -4KV and remains steady.

To me, this looks like a HV breakdown in the multiplier.
It seems that these multipliers are rather expensive at Qservice etc. so I'm considering buying one second hand locally or building one myself. Apart from obvious things like HV isolation (air), >10KV C's and >7KV(?) diodes, would there be any special things to consider when building a multiplier myself?

Any thoughts at all on this?

Thanks!

Raymond


Bob Albert
 

Are you sure the CRT is okay?

Bob


--- On Tue, 4/23/13, raymonddompfrank wrote:

From: raymonddompfrank
Subject: [TekScopes] HV Multiplier defective in 7603?
To: TekScopes@...
Date: Tuesday, April 23, 2013, 10:36 PM

 

Hi all,
My recently restored and very nice early Heerenveen (NL-) made 7603 has developed a problem:
After working OK for a few days, suddenly brightness and focus deteriorated, came back again, deteriorated again and finally the screen became completely dark, all this within 5 minutes. After I opened the case, it immediately became apparent that the HV cage was quite hot, I guess around 50C. Most likely, that's because of Q1216 and Q1218 heating up. No burning smell.

Looking for the fault, being careful and full of respect for HV, I did the following:

1. Disconnected HV cage and measured all LV voltages (-50V, -15V, +5V, +15V, +5-V, +130V), using a 'scope with DVM). All ok, ripple very low.

2. Reconnected HV cage. LV's remain ok, HV Test point (should be -2975V)reaches ca -1.8KV, next goes further negative, something like -2.5KV, then back to around -1.8KV etc. with a period of 1 or 2 seconds. Sort of tick mode below -1.8KV. LV's remain OK, so the 'scope isn't in tick mode. PDA voltage sort of follows change of HV (used P6015). No brightness.

3. Unplugged and discharged post deflection anode. 'Scope behaved as in 2.

4. Disconnected HV Multiplier (U1230) at its input. HV test point reaches almost -4KV and remains steady.

To me, this looks like a HV breakdown in the multiplier.
It seems that these multipliers are rather expensive at Qservice etc. so I'm considering buying one second hand locally or building one myself. Apart from obvious things like HV isolation (air), >10KV C's and >7KV(?) diodes, would there be any special things to consider when building a multiplier myself?

Any thoughts at all on this?

Thanks!

Raymond



raymonddompfrank <r.domp.frank@...>
 

Hi Bob, thanks for responding.
Strictly speaking I'm not sure the CRT is ok; I haven't tested it. I did check isolation resistances (with ohm meter, low voltage) between filament, cathode, grid one and PDA. All resistances >= 2MOhm so there doesn't seem to be a short between any, also when 'scope is switched on.
-4KV is present but "tick mode" starts when connecting HV multiplier input, with or without PDA connected.

Raymond

--- In TekScopes@yahoogroups.com, Bob Albert <bob91343@...> wrote:

Are you sure the CRT is okay?

Bob

--- On Tue, 4/23/13, raymonddompfrank <r.domp.frank@...> wrote:

From: raymonddompfrank <r.domp.frank@...>
Subject: [TekScopes] HV Multiplier defective in 7603?
To: TekScopes@yahoogroups.com
Date: Tuesday, April 23, 2013, 10:36 PM
















 









Hi all,

My recently restored and very nice early Heerenveen (NL-) made 7603 has developed a problem:

After working OK for a few days, suddenly brightness and focus deteriorated, came back again, deteriorated again and finally the screen became completely dark, all this within 5 minutes. After I opened the case, it immediately became apparent that the HV cage was quite hot, I guess around 50C. Most likely, that's because of Q1216 and Q1218 heating up. No burning smell.



Looking for the fault, being careful and full of respect for HV, I did the following:



1. Disconnected HV cage and measured all LV voltages (-50V, -15V, +5V, +15V, +5-V, +130V), using a 'scope with DVM). All ok, ripple very low.



2. Reconnected HV cage. LV's remain ok, HV Test point (should be -2975V)reaches ca -1.8KV, next goes further negative, something like -2.5KV, then back to around -1.8KV etc. with a period of 1 or 2 seconds. Sort of tick mode below -1.8KV. LV's remain OK, so the 'scope isn't in tick mode. PDA voltage sort of follows change of HV (used P6015). No brightness.



3. Unplugged and discharged post deflection anode. 'Scope behaved as in 2.



4. Disconnected HV Multiplier (U1230) at its input. HV test point reaches almost -4KV and remains steady.



To me, this looks like a HV breakdown in the multiplier.

It seems that these multipliers are rather expensive at Qservice etc. so I'm considering buying one second hand locally or building one myself. Apart from obvious things like HV isolation (air), >10KV C's and >7KV(?) diodes, would there be any special things to consider when building a multiplier myself?



Any thoughts at all on this?



Thanks!



Raymond


 

Concerning your point 4, with the presumably bad high voltage
multiplier disconnected, shouldn't the high voltage test point settle
at -2975 volts?

That makes me think something failed in the regulation circuit built
around Q1201, Q1206, and Q1214 including feedback divider R1245.

If the high voltage regulator ran away then maybe it damaged the high
voltage multiplier instead of the reverse.

As to building one with discrete parts, that is what I would do. The
esoteric thing that comes to mind other than standard high voltage
construction methods is the use of a vacuum to remove air from any
potting mixture. I would build the circuit supported in air by its
three external connections without using perforated board and then pot
it.

On Wed, 24 Apr 2013 05:36:25 -0000, "raymonddompfrank"
<r.domp.frank@wxs.nl> wrote:

Hi all,
My recently restored and very nice early Heerenveen (NL-) made 7603 has developed a problem:
After working OK for a few days, suddenly brightness and focus deteriorated, came back again, deteriorated again and finally the screen became completely dark, all this within 5 minutes. After I opened the case, it immediately became apparent that the HV cage was quite hot, I guess around 50C. Most likely, that's because of Q1216 and Q1218 heating up. No burning smell.

Looking for the fault, being careful and full of respect for HV, I did the following:

1. Disconnected HV cage and measured all LV voltages (-50V, -15V, +5V, +15V, +5-V, +130V), using a 'scope with DVM). All ok, ripple very low.

2. Reconnected HV cage. LV's remain ok, HV Test point (should be -2975V)reaches ca -1.8KV, next goes further negative, something like -2.5KV, then back to around -1.8KV etc. with a period of 1 or 2 seconds. Sort of tick mode below -1.8KV. LV's remain OK, so the 'scope isn't in tick mode. PDA voltage sort of follows change of HV (used P6015). No brightness.

3. Unplugged and discharged post deflection anode. 'Scope behaved as in 2.

4. Disconnected HV Multiplier (U1230) at its input. HV test point reaches almost -4KV and remains steady.

To me, this looks like a HV breakdown in the multiplier.
It seems that these multipliers are rather expensive at Qservice etc. so I'm considering buying one second hand locally or building one myself. Apart from obvious things like HV isolation (air), >10KV C's and >7KV(?) diodes, would there be any special things to consider when building a multiplier myself?

Any thoughts at all on this?

Thanks!

Raymond


Craig Sawyers <c.sawyers@...>
 

Subject: [TekScopes] HV Multiplier defective in 7603?
When the multiplier went in my 7603 the trace instantly disappeared. The
sort of intermittent behaviour that you are having kind of indicates a fault
elsewhere in the CRT circuits.

Craig


raymonddompfrank <r.domp.frank@...>
 

Hi Craig,
I have no experience with failing HV multipliers so thanks for this info. Will keep that in mind.

Raymond

--- In TekScopes@yahoogroups.com, "Craig Sawyers" <c.sawyers@...> wrote:

Subject: [TekScopes] HV Multiplier defective in 7603?
When the multiplier went in my 7603 the trace instantly disappeared. The
sort of intermittent behaviour that you are having kind of indicates a fault
elsewhere in the CRT circuits.

Craig


raymonddompfrank <r.domp.frank@...>
 

David,
I thought that the multiplier was needed to allow regulation to work, so -4KV would be reasonable without any load.
Of course, load by a healthy multiplier and CRT would be minimal, so you maybe right, also re. destruction of the multiplier as a consequence of the -4KV.
I will take a look at the feedback/regulation circuit.

Thanks for helping me on my way.

With regard to encapsulating the discrete HV components: What you describe is what I had in mind once the circuit would work "supended in the air", plus thanks for reminding me to get rid of air bubbles.

Raymond

--- In TekScopes@yahoogroups.com, David <davidwhess@...> wrote:

Concerning your point 4, with the presumably bad high voltage
multiplier disconnected, shouldn't the high voltage test point settle
at -2975 volts?

That makes me think something failed in the regulation circuit built
around Q1201, Q1206, and Q1214 including feedback divider R1245.

If the high voltage regulator ran away then maybe it damaged the high
voltage multiplier instead of the reverse.

As to building one with discrete parts, that is what I would do. The
esoteric thing that comes to mind other than standard high voltage
construction methods is the use of a vacuum to remove air from any
potting mixture. I would build the circuit supported in air by its
three external connections without using perforated board and then pot
it.

On Wed, 24 Apr 2013 05:36:25 -0000, "raymonddompfrank"
<r.domp.frank@...> wrote:

Hi all,
My recently restored and very nice early Heerenveen (NL-) made 7603 has developed a problem:
After working OK for a few days, suddenly brightness and focus deteriorated, came back again, deteriorated again and finally the screen became completely dark, all this within 5 minutes. After I opened the case, it immediately became apparent that the HV cage was quite hot, I guess around 50C. Most likely, that's because of Q1216 and Q1218 heating up. No burning smell.

Looking for the fault, being careful and full of respect for HV, I did the following:

1. Disconnected HV cage and measured all LV voltages (-50V, -15V, +5V, +15V, +5-V, +130V), using a 'scope with DVM). All ok, ripple very low.

2. Reconnected HV cage. LV's remain ok, HV Test point (should be -2975V)reaches ca -1.8KV, next goes further negative, something like -2.5KV, then back to around -1.8KV etc. with a period of 1 or 2 seconds. Sort of tick mode below -1.8KV. LV's remain OK, so the 'scope isn't in tick mode. PDA voltage sort of follows change of HV (used P6015). No brightness.

3. Unplugged and discharged post deflection anode. 'Scope behaved as in 2.

4. Disconnected HV Multiplier (U1230) at its input. HV test point reaches almost -4KV and remains steady.

To me, this looks like a HV breakdown in the multiplier.
It seems that these multipliers are rather expensive at Qservice etc. so I'm considering buying one second hand locally or building one myself. Apart from obvious things like HV isolation (air), >10KV C's and >7KV(?) diodes, would there be any special things to consider when building a multiplier myself?

Any thoughts at all on this?

Thanks!

Raymond


Albert <aodiversen@...>
 

Hi Raymond,

For how long did you measure that -4kV? During startup the multiplier puts (considerable?) load on the oscillator since the internal caps and the CRT anode have to be charged. Without that load the HV can increase faster. It might be that the oscillator tube heats up quicker than the feedback tube(s), leading to excessive HV until the feedback comes in.

Albert

David,
I thought that the multiplier was needed to allow regulation to work, so -4KV would be reasonable without any load.
Of course, load by a healthy multiplier and CRT would be minimal, so you maybe right, also re. destruction of the multiplier as a consequence of the -4KV.
I will take a look at the feedback/regulation circuit.

Thanks for helping me on my way.
---
Raymond

--- In TekScopes@yahoogroups.com, David <davidwhess@> wrote:

Concerning your point 4, with the presumably bad high voltage
multiplier disconnected, shouldn't the high voltage test point settle
at -2975 volts?


 

There are no tubes other than the CRT in the 7603. The high voltage
is regulated from the cathode voltage test point so except for
overshoot in the feedback loop, it should never exceed, um, -2975
volts in this case. Excessive current through the high voltage
multiplier could pull it down (toward zero) but never pull it up.

On Wed, 24 Apr 2013 18:33:30 -0000, "Albert" <aodiversen@online.nl>
wrote:

Hi Raymond,

For how long did you measure that -4kV? During startup the multiplier puts (considerable?) load on the oscillator since the internal caps and the CRT anode have to be charged. Without that load the HV can increase faster. It might be that the oscillator tube heats up quicker than the feedback tube(s), leading to excessive HV until the feedback comes in.

Albert


David,
I thought that the multiplier was needed to allow regulation to work, so -4KV would be reasonable without any load.
Of course, load by a healthy multiplier and CRT would be minimal, so you maybe right, also re. destruction of the multiplier as a consequence of the -4KV.
I will take a look at the feedback/regulation circuit.

Thanks for helping me on my way.
---
Raymond

--- In TekScopes@yahoogroups.com, David <davidwhess@> wrote:

Concerning your point 4, with the presumably bad high voltage
multiplier disconnected, shouldn't the high voltage test point settle
at -2975 volts?


 

I will not say that is impossible but if I designed it, the cathode
voltage regulation would function correctly with or without the high
voltage multiplier attached. I think the load on the 6.3 volt AC
heater supply which also comes from T1225 would overwhelm any effect
from the high voltage multiplier load.

You might want to check that the CRT heater is not open. I think you
said earlier it is not shorted to the cathode.

The error amplifier design Tektronix used seems rather unwieldy to me.
I wish they had marked the typical operating voltages on the schematic
but maybe there was too much variation for that to be useful.

The 7613 and 7623 use the same basic design while the 7623A and 7633
use a single ended oscillator driving the transformer instead of a
push-pull oscillator. I looked over the schematics for all four of
those and none provided any insight into the typical operating values.

On Wed, 24 Apr 2013 17:11:45 -0000, "raymonddompfrank"
<r.domp.frank@wxs.nl> wrote:

David,
I thought that the multiplier was needed to allow regulation to work, so -4KV would be reasonable without any load.
Of course, load by a healthy multiplier and CRT would be minimal, so you maybe right, also re. destruction of the multiplier as a consequence of the -4KV.
I will take a look at the feedback/regulation circuit.

Thanks for helping me on my way.

With regard to encapsulating the discrete HV components: What you describe is what I had in mind once the circuit would work "supended in the air", plus thanks for reminding me to get rid of air bubbles.

Raymond

--- In TekScopes@yahoogroups.com, David <davidwhess@...> wrote:

Concerning your point 4, with the presumably bad high voltage
multiplier disconnected, shouldn't the high voltage test point settle
at -2975 volts?

That makes me think something failed in the regulation circuit built
around Q1201, Q1206, and Q1214 including feedback divider R1245.

If the high voltage regulator ran away then maybe it damaged the high
voltage multiplier instead of the reverse.

As to building one with discrete parts, that is what I would do. The
esoteric thing that comes to mind other than standard high voltage
construction methods is the use of a vacuum to remove air from any
potting mixture. I would build the circuit supported in air by its
three external connections without using perforated board and then pot
it.


Albert <aodiversen@...>
 

You're right, I have been busy with 5xx 'scopes lately and simply forgot that the 76xx is further transistorized.
Albert

--- In TekScopes@yahoogroups.com, David <davidwhess@...> wrote:

There are no tubes other than the CRT in the 7603. The high voltage
is regulated from the cathode voltage test point so except for
overshoot in the feedback loop, it should never exceed, um, -2975
volts in this case. Excessive current through the high voltage
multiplier could pull it down (toward zero) but never pull it up.


Hi Raymond,

For how long did you measure that -4kV? During startup the multiplier puts (considerable?) load on the oscillator since the internal caps and the CRT anode have to be charged. Without that load the HV can increase faster. It might be that the oscillator tube heats up quicker than the feedback tube(s), leading to excessive HV until the feedback comes in.

Albert


raymonddompfrank <r.domp.frank@...>
 

My idea about loading the circuit came from the thought that the multiplier input would see a peak-detected DC voltage that had to see some load to discharge. Nonsense, of course. My excuse is that it was at 7am this morning and I had been working all day and night (not on this problem).
Just pulled Q1201. Cathode voltage goes down to far below 2KV so Q1206, Q1214 and associated circuitry could well be ok. Looking at the circuitry driving Q1201 now (R1245 etc). R1245B could be open but I can't easily measure 24.5Mohm.

Next would be the Grid Bias circuitry.

Cathode filament is not open (2 Ohm across P1275 pins 1, 2).

--- In TekScopes@yahoogroups.com, David <davidwhess@...> wrote:

I will not say that is impossible but if I designed it, the cathode
voltage regulation would function correctly with or without the high
voltage multiplier attached. I think the load on the 6.3 volt AC
heater supply which also comes from T1225 would overwhelm any effect
from the high voltage multiplier load.

You might want to check that the CRT heater is not open. I think you
said earlier it is not shorted to the cathode.

The error amplifier design Tektronix used seems rather unwieldy to me.
I wish they had marked the typical operating voltages on the schematic
but maybe there was too much variation for that to be useful.

The 7613 and 7623 use the same basic design while the 7623A and 7633
use a single ended oscillator driving the transformer instead of a
push-pull oscillator. I looked over the schematics for all four of
those and none provided any insight into the typical operating values.

On Wed, 24 Apr 2013 17:11:45 -0000, "raymonddompfrank"
<r.domp.frank@...> wrote:

David,
I thought that the multiplier was needed to allow regulation to work, so -4KV would be reasonable without any load.
Of course, load by a healthy multiplier and CRT would be minimal, so you maybe right, also re. destruction of the multiplier as a consequence of the -4KV.
I will take a look at the feedback/regulation circuit.

Thanks for helping me on my way.

With regard to encapsulating the discrete HV components: What you describe is what I had in mind once the circuit would work "supended in the air", plus thanks for reminding me to get rid of air bubbles.

Raymond

--- In TekScopes@yahoogroups.com, David <davidwhess@> wrote:

Concerning your point 4, with the presumably bad high voltage
multiplier disconnected, shouldn't the high voltage test point settle
at -2975 volts?

That makes me think something failed in the regulation circuit built
around Q1201, Q1206, and Q1214 including feedback divider R1245.

If the high voltage regulator ran away then maybe it damaged the high
voltage multiplier instead of the reverse.

As to building one with discrete parts, that is what I would do. The
esoteric thing that comes to mind other than standard high voltage
construction methods is the use of a vacuum to remove air from any
potting mixture. I would build the circuit supported in air by its
three external connections without using perforated board and then pot
it.


raymonddompfrank <r.domp.frank@...>
 

I thought it had to be something like that, Albert..

Raymond

--- In TekScopes@yahoogroups.com, "Albert" <aodiversen@...> wrote:

You're right, I have been busy with 5xx 'scopes lately and simply forgot that the 76xx is further transistorized.
Albert

--- In TekScopes@yahoogroups.com, David <davidwhess@> wrote:

There are no tubes other than the CRT in the 7603. The high voltage
is regulated from the cathode voltage test point so except for
overshoot in the feedback loop, it should never exceed, um, -2975
volts in this case. Excessive current through the high voltage
multiplier could pull it down (toward zero) but never pull it up.


Hi Raymond,

For how long did you measure that -4kV? During startup the multiplier puts (considerable?) load on the oscillator since the internal caps and the CRT anode have to be charged. Without that load the HV can increase faster. It might be that the oscillator tube heats up quicker than the feedback tube(s), leading to excessive HV until the feedback comes in.

Albert


 

Both the high voltage multiplier and the cathode supply work off of
the peak voltage from T1225 but since the regulator only uses the
rectified and filtered cathode supply for feedback, it does not care
what the high voltage multiplier is doing unless it draws too much
power.

The high voltage and cathode voltage outputs are actually designed to
track pretty closely so the regulator indirectly controls the
acceleration voltage by regulating the cathode voltage because the
former needs to be stable for proper horizontal and vertical
calibration. Thinking about it now, I wonder if CR1244 is there to
roughly temperature compensate the diode voltage drops through the
high voltage multiplier. As the temperature goes up, the cathode
voltage will drop by roughly 120mV/Cdegrees (59.5 x 2 mV/Cdegrees)
which would be about 30mV/Cdegrees for each diode in the multiplier. I
am not sure that the post deflection accelerator voltage needs that
much accuracy but it would explain CR1244.

I would be really suspicious of Q1201. It only operates with
microamps of collector current so if it was suffering from high
leakage, that would explain the high voltage.

Another easy check would be the anode of CR1244 which should be +50.6
volts.

I just noticed that there is no cathode voltage adjustment. The 4
slot 7000 mainframes have a trimmer to adjust the cathode voltage.

On Wed, 24 Apr 2013 20:30:48 -0000, "raymonddompfrank"
<r.domp.frank@wxs.nl> wrote:

My idea about loading the circuit came from the thought that the multiplier input would see a peak-detected DC voltage that had to see some load to discharge. Nonsense, of course. My excuse is that it was at 7am this morning and I had been working all day and night (not on this problem).
Just pulled Q1201. Cathode voltage goes down to far below 2KV so Q1206, Q1214 and associated circuitry could well be ok. Looking at the circuitry driving Q1201 now (R1245 etc). R1245B could be open but I can't easily measure 24.5Mohm.

Next would be the Grid Bias circuitry.

Cathode filament is not open (2 Ohm across P1275 pins 1, 2).


--- In TekScopes@yahoogroups.com, David <davidwhess@...> wrote:

I will not say that is impossible but if I designed it, the cathode
voltage regulation would function correctly with or without the high
voltage multiplier attached. I think the load on the 6.3 volt AC
heater supply which also comes from T1225 would overwhelm any effect
from the high voltage multiplier load.

You might want to check that the CRT heater is not open. I think you
said earlier it is not shorted to the cathode.

The error amplifier design Tektronix used seems rather unwieldy to me.
I wish they had marked the typical operating voltages on the schematic
but maybe there was too much variation for that to be useful.

The 7613 and 7623 use the same basic design while the 7623A and 7633
use a single ended oscillator driving the transformer instead of a
push-pull oscillator. I looked over the schematics for all four of
those and none provided any insight into the typical operating values.

On Wed, 24 Apr 2013 17:11:45 -0000, "raymonddompfrank"
<r.domp.frank@...> wrote:

David,
I thought that the multiplier was needed to allow regulation to work, so -4KV would be reasonable without any load.
Of course, load by a healthy multiplier and CRT would be minimal, so you maybe right, also re. destruction of the multiplier as a consequence of the -4KV.
I will take a look at the feedback/regulation circuit.

Thanks for helping me on my way.

With regard to encapsulating the discrete HV components: What you describe is what I had in mind once the circuit would work "supended in the air", plus thanks for reminding me to get rid of air bubbles.

Raymond

--- In TekScopes@yahoogroups.com, David <davidwhess@> wrote:

Concerning your point 4, with the presumably bad high voltage
multiplier disconnected, shouldn't the high voltage test point settle
at -2975 volts?

That makes me think something failed in the regulation circuit built
around Q1201, Q1206, and Q1214 including feedback divider R1245.

If the high voltage regulator ran away then maybe it damaged the high
voltage multiplier instead of the reverse.

As to building one with discrete parts, that is what I would do. The
esoteric thing that comes to mind other than standard high voltage
construction methods is the use of a vacuum to remove air from any
potting mixture. I would build the circuit supported in air by its
three external connections without using perforated board and then pot
it.


raymonddompfrank <r.domp.frank@...>
 

David,
I had already replaced Q1201 after operating without it reduced cathode voltage a lot. No difference.

Also checked CR1244 and its (cathode) connection to +50V. Looks OK.
That's why I'd like to check R1245B as mentioned earlier but I have no easy way to measure directly. Guess I'll try putting something in parallel to get below 20MOhm, my meter's max. Guess I'll unsolder R2445 (thick film), 6 pins. I have the right equipment to pull solder away.

--- In TekScopes@yahoogroups.com, David <davidwhess@...> wrote:

Both the high voltage multiplier and the cathode supply work off of
the peak voltage from T1225 but since the regulator only uses the
rectified and filtered cathode supply for feedback, it does not care
what the high voltage multiplier is doing unless it draws too much
power.

The high voltage and cathode voltage outputs are actually designed to
track pretty closely so the regulator indirectly controls the
acceleration voltage by regulating the cathode voltage because the
former needs to be stable for proper horizontal and vertical
calibration. Thinking about it now, I wonder if CR1244 is there to
roughly temperature compensate the diode voltage drops through the
high voltage multiplier. As the temperature goes up, the cathode
voltage will drop by roughly 120mV/Cdegrees (59.5 x 2 mV/Cdegrees)
which would be about 30mV/Cdegrees for each diode in the multiplier. I
am not sure that the post deflection accelerator voltage needs that
much accuracy but it would explain CR1244.

I would be really suspicious of Q1201. It only operates with
microamps of collector current so if it was suffering from high
leakage, that would explain the high voltage.

Another easy check would be the anode of CR1244 which should be +50.6
volts.

I just noticed that there is no cathode voltage adjustment. The 4
slot 7000 mainframes have a trimmer to adjust the cathode voltage.

On Wed, 24 Apr 2013 20:30:48 -0000, "raymonddompfrank"
<r.domp.frank@...> wrote:

My idea about loading the circuit came from the thought that the multiplier input would see a peak-detected DC voltage that had to see some load to discharge. Nonsense, of course. My excuse is that it was at 7am this morning and I had been working all day and night (not on this problem).
Just pulled Q1201. Cathode voltage goes down to far below 2KV so Q1206, Q1214 and associated circuitry could well be ok. Looking at the circuitry driving Q1201 now (R1245 etc). R1245B could be open but I can't easily measure 24.5Mohm.

Next would be the Grid Bias circuitry.

Cathode filament is not open (2 Ohm across P1275 pins 1, 2).


--- In TekScopes@yahoogroups.com, David <davidwhess@> wrote:

I will not say that is impossible but if I designed it, the cathode
voltage regulation would function correctly with or without the high
voltage multiplier attached. I think the load on the 6.3 volt AC
heater supply which also comes from T1225 would overwhelm any effect
from the high voltage multiplier load.

You might want to check that the CRT heater is not open. I think you
said earlier it is not shorted to the cathode.

The error amplifier design Tektronix used seems rather unwieldy to me.
I wish they had marked the typical operating voltages on the schematic
but maybe there was too much variation for that to be useful.

The 7613 and 7623 use the same basic design while the 7623A and 7633
use a single ended oscillator driving the transformer instead of a
push-pull oscillator. I looked over the schematics for all four of
those and none provided any insight into the typical operating values.

On Wed, 24 Apr 2013 17:11:45 -0000, "raymonddompfrank"
<r.domp.frank@> wrote:

David,
I thought that the multiplier was needed to allow regulation to work, so -4KV would be reasonable without any load.
Of course, load by a healthy multiplier and CRT would be minimal, so you maybe right, also re. destruction of the multiplier as a consequence of the -4KV.
I will take a look at the feedback/regulation circuit.

Thanks for helping me on my way.

With regard to encapsulating the discrete HV components: What you describe is what I had in mind once the circuit would work "supended in the air", plus thanks for reminding me to get rid of air bubbles.

Raymond

--- In TekScopes@yahoogroups.com, David <davidwhess@> wrote:

Concerning your point 4, with the presumably bad high voltage
multiplier disconnected, shouldn't the high voltage test point settle
at -2975 volts?

That makes me think something failed in the regulation circuit built
around Q1201, Q1206, and Q1214 including feedback divider R1245.

If the high voltage regulator ran away then maybe it damaged the high
voltage multiplier instead of the reverse.

As to building one with discrete parts, that is what I would do. The
esoteric thing that comes to mind other than standard high voltage
construction methods is the use of a vacuum to remove air from any
potting mixture. I would build the circuit supported in air by its
three external connections without using perforated board and then pot
it.


Don Black <donald_black@...>
 

You can measure the 24.5 M resistor by putting a lower value resistor in parallel and then measuring. If it's completely open you'll just get the value of the parallel resistor but if it's OK the resistance will be lower. Then you can calculate the value of the 24.5, there are little programs on the net to do it for you. I'd suggest using a 10 Meg in parallel if your meter can go that high, otherwise a 1 Meg.

Don Black.

On 25-Apr-13 6:30 AM, raymonddompfrank wrote:
 

My idea about loading the circuit came from the thought that the multiplier input would see a peak-detected DC voltage that had to see some load to discharge. Nonsense, of course. My excuse is that it was at 7am this morning and I had been working all day and night (not on this problem).
Just pulled Q1201. Cathode voltage goes down to far below 2KV so Q1206, Q1214 and associated circuitry could well be ok. Looking at the circuitry driving Q1201 now (R1245 etc). R1245B could be open but I can't easily measure 24.5Mohm.

Next would be the Grid Bias circuitry.

Cathode filament is not open (2 Ohm across P1275 pins 1, 2).

--- In TekScopes@..., David wrote:
>
> I will not say that is impossible but if I designed it, the cathode
> voltage regulation would function correctly with or without the high
> voltage multiplier attached. I think the load on the 6.3 volt AC
> heater supply which also comes from T1225 would overwhelm any effect
> from the high voltage multiplier load.
>
> You might want to check that the CRT heater is not open. I think you
> said earlier it is not shorted to the cathode.
>
> The error amplifier design Tektronix used seems rather unwieldy to me.
> I wish they had marked the typical operating voltages on the schematic
> but maybe there was too much variation for that to be useful.
>
> The 7613 and 7623 use the same basic design while the 7623A and 7633
> use a single ended oscillator driving the transformer instead of a
> push-pull oscillator. I looked over the schematics for all four of
> those and none provided any insight into the typical operating values.
>
> On Wed, 24 Apr 2013 17:11:45 -0000, "raymonddompfrank"
> wrote:
>
> >David,
> >I thought that the multiplier was needed to allow regulation to work, so -4KV would be reasonable without any load.
> >Of course, load by a healthy multiplier and CRT would be minimal, so you maybe right, also re. destruction of the multiplier as a consequence of the -4KV.
> >I will take a look at the feedback/regulation circuit.
> >
> >Thanks for helping me on my way.
> >
> >With regard to encapsulating the discrete HV components: What you describe is what I had in mind once the circuit would work "supended in the air", plus thanks for reminding me to get rid of air bubbles.
> >
> >Raymond
> >
> >--- In TekScopes@..., David wrote:
> >>
> >> Concerning your point 4, with the presumably bad high voltage
> >> multiplier disconnected, shouldn't the high voltage test point settle
> >> at -2975 volts?
> >>
> >> That makes me think something failed in the regulation circuit built
> >> around Q1201, Q1206, and Q1214 including feedback divider R1245.
> >>
> >> If the high voltage regulator ran away then maybe it damaged the high
> >> voltage multiplier instead of the reverse.
> >>
> >> As to building one with discrete parts, that is what I would do. The
> >> esoteric thing that comes to mind other than standard high voltage
> >> construction methods is the use of a vacuum to remove air from any
> >> potting mixture. I would build the circuit supported in air by its
> >> three external connections without using perforated board and then pot
> >> it.
>



raymonddompfrank <r.domp.frank@...>
 

Thanks, Don, that's what I had in mind and I have already been looking for a suitable R until I just realized that there are several on this very same thick film device... Should be an easy thing if they're all ok.

Raymond

--- In TekScopes@yahoogroups.com, Don Black <donald_black@...> wrote:

You can measure the 24.5 M resistor by putting a lower value resistor in
parallel and then measuring. If it's completely open you'll just get the
value of the parallel resistor but if it's OK the resistance will be
lower. Then you can calculate the value of the 24.5, there are little
programs on the net to do it for you. I'd suggest using a 10 Meg in
parallel if your meter can go that high, otherwise a 1 Meg.

Don Black.

On 25-Apr-13 6:30 AM, raymonddompfrank wrote:

My idea about loading the circuit came from the thought that the
multiplier input would see a peak-detected DC voltage that had to see
some load to discharge. Nonsense, of course. My excuse is that it was
at 7am this morning and I had been working all day and night (not on
this problem).
Just pulled Q1201. Cathode voltage goes down to far below 2KV so
Q1206, Q1214 and associated circuitry could well be ok. Looking at the
circuitry driving Q1201 now (R1245 etc). R1245B could be open but I
can't easily measure 24.5Mohm.

Next would be the Grid Bias circuitry.

Cathode filament is not open (2 Ohm across P1275 pins 1, 2).

--- In TekScopes@yahoogroups.com <mailto:TekScopes%40yahoogroups.com>,
David <davidwhess@> wrote:

I will not say that is impossible but if I designed it, the cathode
voltage regulation would function correctly with or without the high
voltage multiplier attached. I think the load on the 6.3 volt AC
heater supply which also comes from T1225 would overwhelm any effect
from the high voltage multiplier load.

You might want to check that the CRT heater is not open. I think you
said earlier it is not shorted to the cathode.

The error amplifier design Tektronix used seems rather unwieldy to me.
I wish they had marked the typical operating voltages on the schematic
but maybe there was too much variation for that to be useful.

The 7613 and 7623 use the same basic design while the 7623A and 7633
use a single ended oscillator driving the transformer instead of a
push-pull oscillator. I looked over the schematics for all four of
those and none provided any insight into the typical operating values.

On Wed, 24 Apr 2013 17:11:45 -0000, "raymonddompfrank"
<r.domp.frank@> wrote:

David,
I thought that the multiplier was needed to allow regulation to
work, so -4KV would be reasonable without any load.
Of course, load by a healthy multiplier and CRT would be minimal,
so you maybe right, also re. destruction of the multiplier as a
consequence of the -4KV.
I will take a look at the feedback/regulation circuit.

Thanks for helping me on my way.

With regard to encapsulating the discrete HV components: What you
describe is what I had in mind once the circuit would work "supended
in the air", plus thanks for reminding me to get rid of air bubbles.

Raymond

--- In TekScopes@yahoogroups.com
<mailto:TekScopes%40yahoogroups.com>, David <davidwhess@> wrote:

Concerning your point 4, with the presumably bad high voltage
multiplier disconnected, shouldn't the high voltage test point settle
at -2975 volts?

That makes me think something failed in the regulation circuit built
around Q1201, Q1206, and Q1214 including feedback divider R1245.

If the high voltage regulator ran away then maybe it damaged the high
voltage multiplier instead of the reverse.

As to building one with discrete parts, that is what I would do. The
esoteric thing that comes to mind other than standard high voltage
construction methods is the use of a vacuum to remove air from any
potting mixture. I would build the circuit supported in air by its
three external connections without using perforated board and
then pot
it.


gmail
 

If you needs such resistor, i got one from greece: http://www.qservice-electronics.com
You need to cut the old one into pieces to get it out without damaging the pcb (or have a good desoldering station).
 
In the end it turned out that it wasn't the resistor after all. It was the multiplier that was faulty in my case.
I happen to have a parts 7603, so i could fix it with it's multiplier. Now i'm out of stock, but i have a whole 7603 as spare :)


raymonddompfrank <r.domp.frank@...>
 

Hi Guido,
I have temporarily removed the resistor. I have the right desoldering equipment and already put it in again. It looks fine but all values are over 10% low. Still not sure if that's a problem though.
Are you saying you *have* or you *had* a multiplier available and now have a whole 7603 minus multiplier as spare?
If you have one available, I'm in NL like you...

Raymond

--- In TekScopes@yahoogroups.com, "gmail" <guidozonderspam@...> wrote:

If you needs such resistor, i got one from greece: http://www.qservice-electronics.com
You need to cut the old one into pieces to get it out without damaging the pcb (or have a good desoldering station).

In the end it turned out that it wasn't the resistor after all. It was the multiplier that was faulty in my case.
I happen to have a parts 7603, so i could fix it with it's multiplier. Now i'm out of stock, but i have a whole 7603 as spare :)