Tek 2230, almost fixed but with a problem out of my competences


Pitpat
 

Hi y'all! As you will notice English isn't my first language so if I'll end up being unclear please tell me!
So I recently got this big boy, it runs flawlessly except for the fact that in A horizontal mode the digital readout seems to be stuck on 2s Sec/Div (I'll try to upload some photos)
The strange fact is that the problem does not persist in B mode Hor. mode and does not affect the analog trace.
What board is responsible for the digital readout?
Am i doing something wrong?
In digital mode the trace seems to be present but with the time base stuck at 2 Sec/Div...


Pitpat
 

I've put some photos in here https://groups.io/g/TekScopes/album?id=264522
Thanks!!


Albert Otten
 

Hi Pitpat,

It's not entirely clear to me what is working correctly and what not. The first photos in https://groups.io/g/TekScopes/album?id=264522 show non-store mode with incorrect readout 2s/div. But what is the input signal? With a suitable frequency signal you can show us that the actual time base speed is correct.
The final pair of photo's shows Store mode. Why did you choose 4k mode and Peak detect? What is the reason that we see no trace(s) at all? The manual mentions that in some settings under Peakdet the trace is suppressed. At my 2232 Peakdet seems to show the upper and lower peak traces always (I did not try all posibilities).
BTW Why is there no display of Trigger Level?
BTW If you don't use GPIB then set the dip switches LON and TON both to "1". That eliminates the bright "SRQ" underline (mid top of CRT).

Albert

On Thu, May 27, 2021 at 09:40 AM, Pitpat wrote:


Hi y'all! As you will notice English isn't my first language so if I'll end up
being unclear please tell me!
So I recently got this big boy, it runs flawlessly except for the fact that in
A horizontal mode the digital readout seems to be stuck on 2s Sec/Div (I'll
try to upload some photos)
The strange fact is that the problem does not persist in B mode Hor. mode and
does not affect the analog trace.
What board is responsible for the digital readout?
Am i doing something wrong?
In digital mode the trace seems to be present but with the time base stuck at
2 Sec/Div...


Pitpat
 

Hi Albert, i'm sorry to get to you only by now, my function generator decided to give up and still refuses to turn on... well! a job for tomorrow i guess!
So, about the photos, I've added 4 new ones 2 digital and 2 analog of the same same signal with the same settigs.
As you can see the analog trace is not affected by the incorrect readout but the digital it does, to me, correct me if im wrong, it looks like in digital A hor. mode a trace is present but with a time base way to slow to get any rasonable shape out of it.
For the 4k its what the manual suggest to use while troubleshooting and for the peak detect i have to replace the cmos battery since it resets at every power-on.
About the GPIB thanks for the tip but ill leave it as it is for now since im working on a arduino iee488 converter project.
Thanks a lot for the time :)
Federico


Albert Otten
 

Hi Frederico, now it's clear what's going on I think and I fully agree with your interpretation. If you manage to repair your function generator a 1 Hz signal or so could show that the sweep time/div is not just "too low" but actually the displayed 2 s/div. I wonder if the problem could be related to that battery problem. It's a hell of a job to follow the schematics from the pdf. I am not familiar with the 2230 and my 2232 internals and responded only because the operating instruction an front layout of these scopes are very similar. Hopefully other members can help you straightforward.
Albert

On Thu, May 27, 2021 at 08:19 PM, Pitpat wrote:


Hi Albert, i'm sorry to get to you only by now, my function generator decided
to give up and still refuses to turn on... well! a job for tomorrow i guess!
So, about the photos, I've added 4 new ones 2 digital and 2 analog of the same
same signal with the same settigs.
As you can see the analog trace is not affected by the incorrect readout but
the digital it does, to me, correct me if im wrong, it looks like in digital A
hor. mode a trace is present but with a time base way to slow to get any
rasonable shape out of it.
For the 4k its what the manual suggest to use while troubleshooting and for
the peak detect i have to replace the cmos battery since it resets at every
power-on.


Pitpat
 

Hi thanks again, I'll try to add a new battery asap.
I guess that I'll have to repair my FG, i wonder if a tone generator would work aswell..
That's a lot
Federico


Pitpat
 

Ok, I've just done a test with a PC tone generator: it kinda works, i can clearly see some sort of trace witch with the cursor i confirmed to be 1Hz...
I've also added a button-coin battery to the CMOS to see if something changes but nothing except that now my config are not volatile :)
Could it be the internal digital clk the problem? As usual I'll add a photo to the album as soon as possible
Thanks
Federico


Ed Breya
 

If the sweep time/div readout is stuck, the analog and digital modes will act differently. In analog mode, the sweep will run according to the actual switch setting, regardless of the readout. In storage mode, the sweep will run according to what the readout code tells it - so it may be at the 2 Sec/div, because that's what it's supposed to do. First check the horizontal switch decoder board and connector, mounted next to the sweep-time switch on the sweep board. It could be as simple as a connector unplugged. If it's working properly, all the unique switch position codes should be delivered to the main/processor board.

Ed


Pitpat
 

Good morning Ed! Thanks a lot.
I had a look to the schematics and I didn't find the boat that you are referring, could it the A-B sweep interface board?


Ed Breya
 

Yes, that sounds about right. I couldn't remember the official name after nearly forty years since I designed it, so had to look at the manual. It's A13, with a bunch of comparators, driving weighted resistor network DACs. The switch positions are encoded in the analog/resistance levels, that get read by an ADC on the processor board. Good luck.

Ed


Pitpat
 

Oh! such an honor to have direction from the designer itself! Thanks a lot!
I've located the board you are referring to and found a table (6-4) with all the voltages that i should see on connector j6421 at the specific sweep settigs.
I'll get to it when ill get back from my daily lessons, and post some results.
Do you know where those voltage are referenced?
How should i interpret incorrect values?
Thanks again
Federico


Pitpat
 

Ok, got some results: I've measured the Ares 1-2 voltages coming out of the comparator TCL374 and they seems to be all wrong...
Few volts off and none that goes near 0 that means that we found a culprit right? I've also replaced C766 since it was already of the circuit for testing.
I've posted the table I'm referring to.
Federico


Pitpat
 

Update, something strange happened! I was playing with the gpio expansion of the scope and an arduino when, with the corner of my eye, I've noticed some changes. Toh! The readout changed from 2s to the correct value... but not for all settings!!
Im now suspecting the power lines...
Any opinions would be appreciated
Thanks
Federico


Albert Otten
 

Hi Federico,
Perhaps related: in your 1 Hz picture, why is the amplitude so low? Apparently there there is a lot of noise (50 Hz ripple?) since the peak limits are relatively so far apart.
Albert


Pitpat
 

Hi Albert!
In reality I kinda expected to have some sorta of 50hz noise since I had to use a tone generated by a PC soundcard.
I guessed that with such low frequencies audio signal might have all kind of artefacts.
Also the signal passed through a cheap jack and an alligator-clip to BNC probe, so yes not the cleanest setup.
You gave an idea though, could the problem be related to some sorta of ripple current trough the -8,6 rail? Should I try to decouple it with a cap.?


Albert Otten
 

Hi Federico,
I see, a sound card at 1 Hz is probably not the most ideal signal source!
Don't you have another scope to check power supplies for ripple and correct values? That's more attractive than adding or changing capacitors "in the wild".
In non-store mode your 1 kHz square wave was correct but the readout showed 2 s/div. But the timing switches at diagram <5> and the outputs of U780 do not determine the X10 multiplier. This suggests a fault in common, could be like you say a power line.
Albert

On Wed, Jun 2, 2021 at 10:59 AM, Pitpat wrote:


Hi Albert!
In reality I kinda expected to have some sorta of 50hz noise since I had to
use a tone generated by a PC soundcard.
I guessed that with such low frequencies audio signal might have all kind of
artefacts.
Also the signal passed through a cheap jack and an alligator-clip to BNC
probe, so yes not the cleanest setup.
You gave an idea though, could the problem be related to some sorta of ripple
current trough the -8,6 rail? Should I try to decouple it with a cap.?


Pitpat
 

Yes i do have another scope but is a 20MHz full analog one... I don't think that it would make a good choice with unrepetitive signals like ripples.
Strangely enough i can multiply the 2s to 20s the time/dive if i press the B suppressor an than pull the X10 mag. It's also shown by the readout!


Albert Otten
 

Ripple from power supplies can be repetitive 50 or 100 Hz (or 60/120), best viewed by setting Line trigger mode. With a SMPS it can also be at the frequency of switching (say 25 kHz or so). Albert

On Wed, Jun 2, 2021 at 01:22 PM, Pitpat wrote:


Yes i do have another scope but is a 20MHz full analog one... I don't think
that it would make a good choice with unrepetitive signals like ripples.
Strangely enough i can multiply the 2s to 20s the time/dive if i press the B
suppressor an than pull the X10 mag. It's also shown by the readout!


Pitpat
 

Good afternoon guys! I've posted some photos of the ripple, 3 of the -8.6 REF RES and one from the 42.8vdc smps (the one with the huge spikes)
What do you think? the -8.6v doesn't look so bad imho.


Pitpat
 

Good morning! Last night I've replaced C767 and C768 and filtered with a 100Hz LC Low pass and the ripple was gone, still nothing.
I'm quite stuck, since the replacemets for the comparators are preatty cheap i'm thinking of replacing them aswell.