开云体育

I was going to calibrate my 7a26 plugins and immediately i found the rise time on channel one is horrible
/g/TekScopes/photo/49370/1?p=Name,,,20,1,0,0

Channel 2 seems fine
/g/TekScopes/photo/49370/0?p=Name,,,20,1,0,0

Any thoughts as what to look at would be appreciated. Thanks guys

On Tue, May 15, 2018 at 09:20 pm, lop pol wrote:

I was going to calibrate my 7a26 plugins and immediately i found the rise time
on channel one is horrible
/g/TekScopes/photo/49370/1?p=Name,,,20,1,0,0

That's the symptom of one of the trimmer capacitors going open-circuit on that channel. I think it's the one labelled 'LF' or 'MF' but I can't remember exactly. It's easy to identify because adjusting it will have no effect.

I've seen exactly that symptom on more than one of my 7A26s. I haven't yet replaced the offending trimmers so can't advise on a suitable part, I'm afraid.

Chris

On Tue, May 15, 2018 at 09:20 pm, lop pol wrote:

Any thoughts as what to look at would be appreciated. Thanks guys

See messages #140760 and #140775

/H?kan

Hi Lop Pol (again),
As others answered, it seems those tiny 5mm trimmer caps are prone to fail.
I had one of such failed on the pre-amplifier of my 464, and had also one failed on the vertical output amp of the 7623A.
I noticed, in both cases (so it doesn't seem to be only coincidence) that they're were too stiff to move when first attempted to calibrate them and then, on applying more force - snap - the tiny screw driver head would turn, but I noticed that in both cases, the ceramic discs were not turning.
And in both cases (the 464 and the 7623A), I managed to use a tiny pliers covered with rubber to force turn the ceramic disk which was stuck (maybe due to dried out lubricant) and was also able to make the screw head "bond" again to the ceramic disc by applying a little bit of flux to the head and re-heating it with a soldering iron.
On the 464, the repair ended-up flaky. It didn't survive and I had to replace the trimmer cap.
On the 7623A, it worked well and it's still there.
Rgrds,
Fabio

On Wed, May 16, 2018 at 08:29 am, Fabio Trevisan wrote:

Hi Lop Pol (again),
As others answered, it seems those tiny 5mm trimmer caps are prone to fail.
I had one of such failed on the pre-amplifier of my 464, and had also one
failed on the vertical output amp of the 7623A.
I noticed, in both cases (so it doesn't seem to be only coincidence) that
they're were too stiff to move when first attempted to calibrate them and
then, on applying more force - snap - the tiny screw driver head would turn,
but I noticed that in both cases, the ceramic discs were not turning.
And in both cases (the 464 and the 7623A), I managed to use a tiny pliers
covered with rubber to force turn the ceramic disk which was stuck (maybe due
to dried out lubricant) and was also able to make the screw head "bond" again
to the ceramic disc by applying a little bit of flux to the head and
re-heating it with a soldering iron.
On the 464, the repair ended-up flaky. It didn't survive and I had to replace
the trimmer cap.
On the 7623A, it worked well and it's still there.
Rgrds,
Fabio

On Tue, May 15, 2018 at 09:20 pm, lop pol wrote:

I was going to calibrate my 7a26 plugins and immediately i found the rise

time

on channel one is horrible
/g/TekScopes/photo/49370/1?p=Name,,,20,1,0,0

Channel 2 seems fine
/g/TekScopes/photo/49370/0?p=Name,,,20,1,0,0

Any thoughts as what to look at would be appreciated. Thanks guys

cmjones01
That's the symptom of one of the trimmer capacitors going open-circuit on that channel. I think it's the one labelled 'LF' or 'MF' but I can't remember exactly. It's easy to identify because adjusting it will have no effect.

I've seen exactly that symptom on more than one of my 7A26s. I haven't yet replaced the offending trimmers so can't advise on a suitable part, I'm afraid.

Chris

zenith5106
4:54 am #147977

On Tue, May 15, 2018 at 09:20 pm, lop pol wrote:


Any thoughts as what to look at would be appreciated. Thanks guys

See messages #140760 and #140775


I replaced the caps /g/TekScopes/photo/49370/2?p=Name,,,20,1,0,0
Now I can calibrate the darn thing. Many thanks guys!!

Messing around today after i THOUGHT i had things straighted out, I was looking at the rise time again and 50mV is good 10mV is fine but now 20 mV only has horrible rise time. Do you guys think that means another cap is open somewhere?

On Thu, May 17, 2018 at 10:16 pm, lop pol wrote:

Messing around today after i THOUGHT i had things straighted out, I was
looking at the rise time again and 50mV is good 10mV is fine but now 20 mV
only has horrible rise time. Do you guys think that means another cap is open
somewhere?

If it's only happening on the 20mV/div setting, it's probably something to do with the attenuator compensation capacitors. These are also prone to problems, though I've found that it's not so much the capacitors that cause trouble as the solder joints to the pins of the attenuator module. They need a careful hand and a fine iron to tidy up.

You won't be able to set the attenuator compensation correctly without an RC normalizer on the input. I made my own with a little box containing a 1Mohm resistor and a trimmer cap (set the trimmer cap like you would a scope probe on a known-good input). Without the normalizer, the attenuator will appear to set up correctly though some adjustments have very little effect, but in fact the result is badly wrong. The adjustments on the attenuator modules interact because they're cascaded, so it's important to follow the procedure in the manual. I found that the 100mV/div setting (the first 10x attenuator) was hard to get right so that the square wave looked right both with and without the normalizer, so I had to make gentle tweaks adding and removing the normalizer to get the best setting.

The point of all this is to get the frequency compensation and input capacitance the same at all attenuator settings.

The procedure I use if I can't be bothered to find the manual is this:
- set attenuator to 10mV/div (straight-through, no attenuation)
- connect fast rise square wave generator to input with in-line 50R terminator (to make sure there are no aberrations due to the cable), *no* RC normalizer
- set up LF/MF/HF trimmers on amplifer board for optimum settings
- insert RC normalizer between square wave gen and input. You can remove the terminator if it makes it easier for your generator to produce the right output, because we no longer need such fast edges for the rest of the adjustments
- adjust normalizer trimmer for optimum square wave
- set attenuator to 20mV/div. Adjust 2x attenuator trimmers for optimum corner and flat top.
- set attenuator to 50mV/div. Adjust 5x attenuator trimmers for optimum corner and flat top.
- set attenuator to 100mV/div. Adjust first 10x attenuator trimmers for optimum corner and flat top.
- check 200mV/div and 500mV/div settings on the way past
- set attenuator to 1V/div. Adjust second 10x attenuator trimmers for optimum corner and flat top.
- check 2V/div and 5V/div settings
- go back and check all attenuator settings again
- remove normalizer
- check all attenuator settings again
- repeat until it's good enough

Don't do this without the screening cover over the attenuators because it makes a significant difference to the adjustments. I wasted enough time with that!

Chris

On Fri, May 18, 2018 at 02:24 am, cmjones01 wrote:

On Thu, May 17, 2018 at 10:16 pm, lop pol wrote:

Messing around today after i THOUGHT i had things straighted out, I was
looking at the rise time again and 50mV is good 10mV is fine but now 20 mV
only has horrible rise time. Do you guys think that means another cap is

open

somewhere?

If it's only happening on the 20mV/div setting, it's probably something to do
with the attenuator compensation capacitors. These are also prone to problems,
though I've found that it's not so much the capacitors that cause trouble as
the solder joints to the pins of the attenuator module. They need a careful
hand and a fine iron to tidy up.

You won't be able to set the attenuator compensation correctly without an RC
normalizer on the input. I made my own with a little box containing a 1Mohm
resistor and a trimmer cap (set the trimmer cap like you would a scope probe
on a known-good input). Without the normalizer, the attenuator will appear to
set up correctly though some adjustments have very little effect, but in fact
the result is badly wrong. The adjustments on the attenuator modules interact
because they're cascaded, so it's important to follow the procedure in the
manual. I found that the 100mV/div setting (the first 10x attenuator) was hard
to get right so that the square wave looked right both with and without the
normalizer, so I had to make gentle tweaks adding and removing the normalizer
to get the best setting.

The point of all this is to get the frequency compensation and input
capacitance the same at all attenuator settings.

The procedure I use if I can't be bothered to find the manual is this:
- set attenuator to 10mV/div (straight-through, no attenuation)
- connect fast rise square wave generator to input with in-line 50R terminator
(to make sure there are no aberrations due to the cable), *no* RC normalizer
- set up LF/MF/HF trimmers on amplifer board for optimum settings
- insert RC normalizer between square wave gen and input. You can remove the
terminator if it makes it easier for your generator to produce the right
output, because we no longer need such fast edges for the rest of the
adjustments
- adjust normalizer trimmer for optimum square wave
- set attenuator to 20mV/div. Adjust 2x attenuator trimmers for optimum corner
and flat top.
- set attenuator to 50mV/div. Adjust 5x attenuator trimmers for optimum corner
and flat top.
- set attenuator to 100mV/div. Adjust first 10x attenuator trimmers for
optimum corner and flat top.
- check 200mV/div and 500mV/div settings on the way past
- set attenuator to 1V/div. Adjust second 10x attenuator trimmers for optimum
corner and flat top.
- check 2V/div and 5V/div settings
- go back and check all attenuator settings again
- remove normalizer
- check all attenuator settings again
- repeat until it's good enough

Don't do this without the screening cover over the attenuators because it
makes a significant difference to the adjustments. I wasted enough time with
that!

Chris

I changed the attenuators with a set of known good ones. Still the same 20mV issue.

20mV /g/TekScopes/photo/49370/4?p=Name,,,20,1,0,0

10mV /g/TekScopes/photo/49370/3?p=Name,,,20,1,0,0

50mV is the same as the 10mV so you get the picture.

Pretty crazy bad.

Hello Lop Pol,

Well... Attenuator blocks are not really swapable "Plug-andPlay" as they are usually adjusted to match each 7A26's input capacitance...
But, since the behavior is consistent and the performance is good on the other attentuator settings, I think you successfully ruled-out the attenuator block as a possible cause.

With all that in perspective, I can only see 3 possible causes for this problem (if logic is sound and if Mr. Murphy is not fooling around your 7A26).
1. A poor leaf switch contact(s) of the X2 attenuation section (the B contacts, as they're named on the schematics), either the one before the X2 attenuator block, or the one after the X2 attenuator block.
2. Or the R106 / LR106, whichever is the case on your particular 7A26.
3. Bad contact (or dry solder joint) at the X2 attenuator block contact pins, but that seems rather unlikely as those are gold plated "collet-type" socket pins and seems very reliable to me.

From all 3, the most likely seems to be number 1, the leaf switch contacts...
The problem is that those contacts are the ones on the back side of the attenuator block PCB (the same side of the attenuator rotary cam which is a lot of work to disassemble).
If the bad contact doesn't clear with contact cleaner (and be sure you use only the ones that are safe to plastics), I`m afraid you have some tough work ahead because, to clean them as per the manual, with IPA and paper, you'll have to dismantle the whole thing.

Rgrds,

Fabio

Fabio. There is no way a simple 33 ohm resistor could cause that... pshhh... HAHA. Nice job!!

/g/TekScopes/photo/49370/7?p=Name,,,20,1,0,0

/g/TekScopes/photo/49370/6?p=Name,,,20,1,0,0

Now I can put the old blocks back in and.... Wait until the next problem rolls around.

Thanks for everyones help again!!!

Lop, great news!
One thing always puzzled me is why Tek insisted with those crappy carbon composite resistors for so long.
Almost into the eighties they were still using those things, while I can't hardly remember having seen carbon composite resistors in any consumer electronics that I was already playing around in the 70s.
I can't remember a single tape-deck or amplifier from the 70s (transistor of course) that would have carbon composite resistors inside!
They were too conservative maybe? Or they had a large stock of Carbon Comp resistors?
I know that in some applications, like H.V., Carbon comp are more robust and had an ability to withstand higher voltages... but everywhere else they just seem to be a poor choice.
I`m glad that it's sorted out and that it was not the leaf contacts... because it would be a hell of a job to clean.
Rgrds,
Fabio

One thing always puzzled me is why Tek insisted with those crappy carbon composite resistors for so
long.
Almost into the eighties they were still using those things, while I can't hardly remember having
seen
carbon composite resistors in any consumer electronics that I was already playing around in the 70s.
I can't remember a single tape-deck or amplifier from the 70s (transistor of course) that would have
carbon composite resistors inside!

Low self inductance. It is a simple carbon-loaded rod. Whereas metal film resistors are usually spiral
cut and have a measurable inductance.

You can easily measure this effect on an RF bridge.

Craig

One thing always puzzled me is why Tek insisted with those crappy carbon composite resistors for so
long.

Allen-Bradley stopped making them in the early '93. Caused us no end of grief at Oxford Instruments
(where I worked at that time), where we used them as low-cost cryogenic temperature sensors, needing
several thousand of a particular value each year.

When AB discontinued them, we bought up as much global stock as we could get our paws on. Given that
they must eventually have run out by now, I have no idea what they currently use.

Craig

On Fri, May 18, 2018 at 09:32 am, lop pol wrote:

Fabio. There is no way a simple 33 ohm resistor could cause that... pshhh...
HAHA. Nice job!!
/g/TekScopes/photo/49370/7?p=Name,,,20,1,0,0
/g/TekScopes/photo/49370/6?p=Name,,,20,1,0,0

Now I can put the old blocks back in and.... Wait until the next problem rolls
around.

Remarkable! I never thought those little resistors could have such an effect, innocent-looking things that they are. Well done for tracking it down, and thank you for posting the result. I'll definitely bear this in mind if I see similar symptoms on my 7A26s.

Chris