Argh! When in doubt, believe the Tek manual. I read that a 2 ohm load was sufficient, resulting in 2.5A draw. Wrong. Added a second 2 ohm on the options line (because the fusing on the 2 exposed +5VD lines is 5A each). PSU now working perfectly. That was a huge waste of time. At least I learned a lot about the PSU. :)

The inner and outer parts of the knob come off together after removing the setscrew.? Under the knob, the pot itself (and an indexing washer) is held in place with a nut.

Bob Weiss

Thank you for the replies.
I am away from the lab until Monday afternoon.
But I am almost sure that the signal at the emitter of Q310 is correct.
Roger

开云体育

I agree checking pin 7 of U350 is a good idea. That said pin 1 of U350A is the output of a unity gain amplifier summing in a DC voltage ?and should not have any steps. It should reflect the same DC voltage as the wiper of the offset multiplier R350. ? ?On the step generator schematic check the emitter of Q310 to see if you have all ?10 steps there.?

It is absolutely important to do CAL 06! You should be able to
see the flaws on any high speed square wave input. If you have
a PG506, it will definitely show the flaws.

You can get most of the way there by doing CAL 06, substituting
the PG506 in fast rise mode for the TD pulser. That is what is
done for the slower scopes.

-Chuck Harris


On Thu, 13 Oct 2022 02:41:49 -0700 "NigelP"
<nigel-pritchard@...> wrote:

I guess my next question is.... is it worth re-doing CAL 06 using the
equipment that I do have rather than accepting nominal values?

In the absence of a TD Pulser, I have the following TM50x modules:-
PG506, PG508, TG501 (non-A) and a HP 8657B (sine wave of course).
Would using any one of those improve the set-up beyond the defaults,
or would it be a pointless excercise?

So for example using the PG506 (as specified in the CAL equipment
list) but WITHOUT the TD pulser would I gain any set-up improvement
over any of the default values?

Hi Nigel,

In Test 02, all of the error codes listed are unique orthogonal values.

I don't recall what happens if more than one switch is stuck, but I am
pretty sure that the scanning matrix used for reading the front panel
switches would fail to provide correct information in that case.

If two switches are stuck, it is likely that the Test 02 routine would
report a completely different switch as being stuck... but if you are
really curious, I guess you could try it by pressing two switches while
you are powering on your scope.

-Chuck Harris



On Thu, 13 Oct 2022 02:56:41 -0700 "NigelP"
<nigel-pritchard@...> wrote:

I guess my issue was not appreciating the mix of digital values and
analog values, something that would never have occured to me; my
brain thinks in either one *or* the other.

Just for the hell of it (and at the risk of boring everybody with too
much detail), how would you interpret the Error Code results for
Table 6-4 in the Service Manual under Test 02? The error codes range
from 01 to 65; are these also summed results of binary values, or is
there another interpretation?

The step generator looks OK, the step amplifier has a problem. Take some more measurements at U350 pins 1 and 7. Are there more steps present there?

Dear All,
I am struggling with this unit for some time.
The waveform at TP305 is correct, according to the manual.
However I can not obtain more than 5 steps at TP-306.
I have checked the voltage at U305B-3 which should "define" the maximum number of steps and it varies from 0.05 to 4.75V , as expected.
Just as a desperate measure I have alredy replaced U305, U220, U350, Q306 and Q324 without success.
I have created a photo album with the waveforms at TP-305, TP-306 , tthe 577 display as well as the pertinent schematics sheets.
See /g/TekScopes2/album?id=279289.
Any help will be welcome.
Thanks,
Roger

Rogerio O <rodd414@...> added the album Tek 577 unable to display more than 5 steps: The waveform at TP305 is fine, but when I turn the knob to increase the number of step s, the maximum I can get is 5. I have attached the schematics so anybody willing to help does not have to look up for them.

I guess my issue was not appreciating the mix of digital values and analog values, something that would never have occured to me; my brain thinks in either one *or* the other.

Just for the hell of it (and at the risk of boring everybody with too much detail), how would you interpret the Error Code results for Table 6-4 in the Service Manual under Test 02? The error codes range from 01 to 65; are these also summed results of binary values, or is there another interpretation?

I guess my next question is.... is it worth re-doing CAL 06 using the equipment that I do have rather than accepting nominal values?

In the absence of a TD Pulser, I have the following TM50x modules:- PG506, PG508, TG501 (non-A) and a HP 8657B (sine wave of course). Would using any one of those improve the set-up beyond the defaults, or would it be a pointless excercise?

So for example using the PG506 (as specified in the CAL equipment list) but WITHOUT the TD pulser would I gain any set-up improvement over any of the default values?

开云体育

In the digital world, you can have all sort of variables.? Bytes (8 bit values), words (can be 16 bit values), etc... Much of the vocabulary depends on the underlying processor.

True and false are generally represented as single values, 1 for true, 0 for false.? You could use an 8 bit variable to store true and false, as well as a 32 bit variable.? Since the system interprets any non-zero value as true, for even an 8 bit variable, you're wasting 254 values....

You can have what are called bit variables, where each bit in a variable is treated as a true or false, so 010110 is false, true, false, true, true, false as an example.?

If each bit in a variable is considered the result of a test, then your composite result is an array of ones and zeros.? If in two tests, the bits assigned have the 8 bit values of 1 (00000001) and 2 (00000010), then a composite word giving the result of those two tests could be 00000000, 00000001, 00000010, or 00000011.? These in decimal are 0, 1, 2, 3.? The word being formed with the logical OR of the error codes, and the error code being the bit number of the failing tests.? Those tests can be thought of tests 1, 2, 4, 8, 16, etc.? Or if you want, the composite error code is the result of the answers "did this test fail" which flips the meaning of the bits.

Hope this helps.? Its been around for a long while, but shows up more in microprocessors, assembly language, and the like where memory can be precious (and not in PC/Linux code where memory is definitely NOT).

Harvey

Hi Nigel,
Thanks for coming back on this. Chuck being Chuck, his explanation was more extensive and clearer than my rather terse remarks. It illustrates exactly what you write as regards wording of a message, even irrespective of mother tongue. Having done business in the Far East and not being a native English speaker myself, I recognize what you're talking about.

Quite often I find that the same questions are asked over and over again, where the answer is readily available by (computer-supported) searching in a group's threads. Chuck, being one of the most knowledgeable people in this group and others, often has the patience to still post a new response. It shows respect for the person asking the question, as does your response for those who answered.

Raymond

Apologies if I've missed the point there; I have to say that despite your inputs and Chuck's it never actually occured to me *precisely* what was written in the specs. It did not at any time occur to me that one had to make a summation of two error codes in order to arrive at the holistic answer.

It was not to say that no-one was giving me the answers, but merely that my understanding was still vague and unclear.

Clearly my brain is analog and not digital, and having to add two digital values to arrive at an otherwise unwritten analog one (I am pretty sure that the number "03" did not appear in any specification/manual I read, but I stand to be corrected) is a bit off-of-centre in my tiny logical world.

Regardless of all that I extend many thanks to all those who have in some way helped me to a successful conclusion (any many others of course). It just goes to show how a subtle change of wording suddenly brings a whole new meaning to life!

As an employee of a multi-national (Semiconductor) company for several decades I had to communicate with hundreds of colleagues for whom English was not their native tongue and I quickly learnt how NOT to write emails!!!! I got to writing, re-writing and re-re-writing many emails (before pushing the SEND button) just to make sure that the recipients would know what I had actually intended to say rather than what I had hastily written assuming they would fully understand my contracted English! The English language can sometimes be remarkably efficient and yet sometimes be missing the critical point of understanding!

On Sun, Oct 9, 2022 at 05:21 PM, NigelP wrote:

At last, someone who actually recognises the error code!!!! Great news Chuck; I had a feeling in my bones that I'd get a response!

Hi Nigel,
The format for these messages is well-documented (SM) and the occurrence of the "03" value has been included in many messages in this and its sister group over time.
As an example, I included it in my message to you in another thread, last july:

/g/TekScopes/message/194621

Raymond

Sorry, got dragged into something else, back on this now. I know about the cap error on the schematics, haven't replaced those 2. Again, it works if the +5VD feedback input to the regulator board is broken, doesn't if connected.

I have checked U1371C. The ref input is correct. The opamp is working. If I disconnect J233 and put the dummy load on the inverter side, PSU starts fine, voltages are fine but a bit high, since there is no feedback. I can then drive the regulator side +5VD with an external supply and see that U1371C pin 8 does the right thing, going positive if the external supply is low, going down towards zero if the external supply is too high. So, that's working fine. But, removing the external supply and reconnecting +5VD from inverter board to regulator board, ticking. Monitoring various bits with a scope, PSU starts, +5VD starts up at about 5.4V. U1371C pin 8 starts decreasing in voltage, which is correct. But, it overshoots too low and the inverter shuts down. This is puzzling because the feedback loop clearly is at least semi-functional. This is a strange beast, the opamp at balance is basically running full gain, the loop being closed by the feedback to the inverter then thru the +5VD line back to the opamp. There an AC feedback loop on the opamp, clearly to provide some damping for the long round trip. Back to being confused. Oh, I also looked at the feedback at pin 8 on the opamp. Goes quickly high when the PSU is turned on, then starts heading down towards 0, exactly as expected. (High means output is too low). But, it never stabilizes at ~1V where it should end up, just keeps going down until the PSU shuts down. Repeat every second or so.

Bill

We have reached a point in time where the tunnel diodes
of old are slower than ordinary SMD logic and comparator
parts.

Leo Bodner has available a pulser that easily exceeds
the rise time specs needed for calibrating the 2467, but
it doesn't have the correct pulse size... as I recall.

I have one I bought for evaluation, but I have never
used it for calibration.

Others may have experience using it for that purpose.

I used to use a tunnel diode pulser, but I switched over
to the CG5010. It makes the whole calibration job easier.

-Chuck Harris


On Tue, 11 Oct 2022 07:18:16 -0700 "NigelP"
<nigel-pritchard@...> wrote:

SUCCESS !!!!!!!

CAL 06 and CAL 07 just performed (without making any adjustments) and
I noticed that the dots that replace the "???????" during the CAL
procedure disappeared (it might have been after the CAL 06). All much
easier than I anticipated.

The scope now boots up as happy as a sand boy into full operational
mode without any silly effects on screen.

It never ocurred to me that the binary error code could be a SUM of
the ones above.

Many thanks for your assistance Chuck; it's great to have experts
around! I can rest my brain now.

Now to think about that pulser..... if I can't find a real TEK item
is there a home-made design that could be built (assuming I can find
a tunnel diode these days.... I'll have to raid my goodies boxes)?