Tek 2465 (1st gen) - help with A5 board troubleshooting?


AngiesHusband
 

Hi Folks,

I am looking for advice from the Tek 2465 gurus here. Tons of great historical info in the forum, and it has been super helpful; but I have finally hit the limits of where my limited troubleshooting skill can take me alone.

The unit is an original series 2465 (SN: B01x vintage). I’ve had great success in re-capping both PSU boards, but now having some irksome issues with the display and delta-T/delta-V controls.

Observed behaviors of the unit (may or may not be relevant):
*Boot time seems slow. At least slower than my other 2465 (SN: B05x vintage) - e.g., the time elapsed from “power on” to “trace displayed”
*Scope seems to "forget" some of the settings when it reboots
*When the trace markings are on (knob CW), there are little “single pixel” specks that flash now and then in the upper and lower margins of the scope.
*When the delta-T/delta-V markers are on - same thing happens, little specks randomly seen on the scope
*The biggest issue: when the "delta ref" or "delta" knob is turned, the marker stutters as it moves. Sometimes even moves randomly in the opposite direction it is being turned. It is the worst with turning the "delta" knob clockwise... Less of an issue if you turn the knob veeeeeery slooooow.


Measurements taken:
*J119 voltages, p-p ripple, and 2x mains ripple are all within spec. I am using the “ground banana plug strap together” note that Chuck Harris recommended.( https://groups.io/g/TekScopes/message/160083 )
*J118-2 voltage adjusted for 2.500 V exactly between full CCW and full CW measurements (e.g., +1.2335V and -1.2565V)
*A5 board Waveform <1> on U2556-4 looks good. 10MHz exactly. A bit more rounded like a sin wave, vs what is shown in the service manual
*Waveform <2> on U2468-5 looks good. Takes some fiddling with holdoff to get it to trigger correctly, but looks like the service manual
*Waveform <3> on U2556-10 is very messy... seems like the basic form is there but lots of jitter and I can’t get it to trigger/stabilize. It also seems to "hiccup" from time to time.
*Waveform <4> on U2034-19 is also messy, in the same way that <3> is but just “upsidedown”
*TP505 is a 1.25MHz square wave with a rounded leading edge at the top.

I started to trace back the problematic waveforms, but after doing that for a bit, I realized I ran myself in a circle (literally) and thought it was time to ask the experts! Any thoughts or recommendations would be greatly appreciated!

Cheers,
Paul


Jean-Paul
 

Please check the A5 NVRAM backup batter or Dallas part, and see the many notes on A5 SMD lytic catastrophie and damages.

Bonjour cher Paul:

Your A5 board my have a dead battery or is loosing the CAL in the NVRAM.

The caps leak and damage the PCB traces.

Many long threads on this topic.

Also check the huge thread on EEVBLOG "tektronix 2465B teardown"

Bon Chance,

Jon


AngiesHusband
 

Hi Jon,
This is actual a first generation 2465. So not a 2465A or 2465B. I do not think the NVRAM or SMD cap issue applies to this unit.

Any other thoughts?

Paul


cmjones01
 

Indeed. The 2465 has an EAROM, not a battery-backed RAM, for storing its
settings and calibration constants.

Chris

On Sat, 13 Mar 2021, 20:42 AngiesHusband via groups.io, <pb77=
pm.me@groups.io> wrote:

Hi Jon,
This is actual a first generation 2465. So not a 2465A or 2465B. I do
not think the NVRAM or SMD cap issue applies to this unit.

Any other thoughts?

Paul






 

The EAROM in the 2465 is Tek part # 156-1566-00 (General Instruments part # EA1400, according to the Common Design Parts Catalog, 1982). They are available on eBay, if you need a replacement (the prices are are around $50 US, which isn't pocket change, but less than dinner out). Lead Electronics seems to also have it in stock for less (no idea what shipping will add to that): https://www.leadelectronics.com/product-details.php?item_no=ER1400&manu_no=GI&catid=97&utm_source=supplyFrame&utm_medium=buyNow

There is also an EAROM replacement board targeted at the arcade console community. It is intended for a different part, but it may point the way toward replacing the now discontinued part with something easier to source: http://www.jrok.com/hardware/earom/

Here is the data sheet for the EA1400: https://www.tautec-electronics.de/Datenblaetter/Schaltkreise/ER1400.pdf
You can also get the GI handbook from BitSavers: http://www.bitsavers.org/components/gi/_dataBooks/1983_GI_Electrically_Alterable_Non-Volatile_Memory_Handbook_1983.pdf


AngiesHusband
 

So is the thought that the EAROM might be bad and causing the bad behavior I am observing in the unit? Are there further tests I should pursue to isolate the EAROM functionality?

Thanks,
Paul


Mark Litwack
 

Keep in mind the 2465 does not store changes to front panel settings right away. It stores them after 10 seconds of no changes to any settings. And the 10 second timer starts only when changes occur to soft settings that are solely tracked by the processor, such as cursor push buttons. Changes to hard switches, such as the horizontal sweep, do not trigger the timer since they can be read again on boot. This is all to reduce write cycles to the EA1400.

In other words, if you change a soft knob or button and turn the scope off in less than 10 seconds, your change won't be there when you turn it back on. Perhaps you're seeing this behavior?

-mark


 

Paul,

Well, my thought was that the EAROM might be bad, but I've never paid attention to whether or not my 2465 kept the values of the soft controls between power cycles, so I wasn't sure if the behavior you reported was normal. It seems like it might not be bad, depending on how quickly you power it off after making a change, but I just ordered one of the EA1400s to have in my parts kit for the 2465, just in case of future need (I ordered the expensive Tek NOS part from eBay, because I'm impetuous. I should have waited till I found the Lead Electronics listing).

The 2465 includes diagnostic firmware that runs when the system is powered on and, in the words of the fine service manual "verify proper operation of much of the instrument's circuitry." You can also run the diagnostic routines manually by holding in the delta-V and delta-t and pressing the SLOPE button. This will present you with a menu of test routines that may be scrolled through using the TRIGGER MODE switch. You select the test routine using the TRIGGER COUPLING switch: up starts the routine, down stops it. If you press TRIGGER COUPLING up while the routine is selected (but not running) you switch to loop mode where the routine will run continuously until stopped. The diagnostic routines are described in the service manual starting on page 6-9.

Test number 04 checks the EAROM for parity errors, checksum errors, and bad read after write, so if you are not seeing a test failure on start up then the EAROM is probably okay.

Do you have a service manual for the scope? If you do not you can download one from TekWiki: https://w140.com/tekwiki/wiki/2465

-- Jeff Dutky


AngiesHusband
 

Ok, I will check that out! The Lead Electronics link seems to be a pretty good price - thanks for the link!

How about thoughts the issues with the display:
*When the trace markings are on (knob CW), there are little “single pixel” specks that flash now and then in the upper and lower margins of the scope.
*When the delta-T/delta-V markers are on - same thing happens, little specks randomly seen on the scope
*The biggest issue: when the "delta ref" or "delta" knob is turned, the marker stutters as it moves. Sometimes even moves randomly in the opposite direction it is being turned.

I can't see that being an EAROM issue... feels more like something with the logic/timing/etc...?

Paul


 

Paul,

I am far from the best person to be making diagnoses of these display issues, but since I happen to also have a 2465 handy, let me try to reproduce your symptoms. Then we might at least know if they are a common malfunction (even if we can't know if they are or are not an actual malfunction).

I powered up my scope, hooked up a P6131 to channel one, connected to the calibrator signal, turned off all but channel 1 (all off), set the time base to 1 ms, and the V/DIV to 100 mV.

When displaying either the delta V or delta T cursors and moving them actively, I occasionally get a flicker of a pixel in the upper and lower readout areas, more frequently in the lower area than in the upper. The flickering pixels share intensity with the readout. The flickering pixels happen whether or not the readout is active, but only happen with the cursors are active (and they have their own read out that does not turn off when the scale factors are turned off).

Does this sound like what you are seeing?

I will try to get pictures that I can post to the photos section.

I don't see any "stuttering" of the delta REF cursor on my scope. I would suspect a bad or dirty rotary encoder. I have no idea what to do about that (again, other folks here will probably have better advice).

-- Jeff Dutky


Chuck Harris <cfharris@...>
 

Contrary to how a modern person might think they are
done, the 2465 (orig) does not use digitizer pots to
handle the variable controls on the front panel. They
use actual pots. The cursors use a very strange two
element, two wiper pot without any rotation stop.

When one wiper is about to run off the end of one pot's
element, the other wiper is just starting on its element.

The CPU figures it all out when the pot's position is
digitized.

Your cursor problem is painfully common, and is caused
by noisy pot elements. It can easily be fixed by taking
the pot apart and putting a drop of Caig's FaderLube on
the element.

I have never seen a failed EAROM on a 2465 (orig).

-Chuck Harris

AngiesHusband via groups.io wrote:

Ok, I will check that out! The Lead Electronics link seems to be a pretty good price - thanks for the link!

How about thoughts the issues with the display:
*When the trace markings are on (knob CW), there are little “single pixel” specks that flash now and then in the upper and lower margins of the scope.
*When the delta-T/delta-V markers are on - same thing happens, little specks randomly seen on the scope
*The biggest issue: when the "delta ref" or "delta" knob is turned, the marker stutters as it moves. Sometimes even moves randomly in the opposite direction it is being turned.

I can't see that being an EAROM issue... feels more like something with the logic/timing/etc...?

Paul






 

Chuck,

I don't think I count as quite a "modern person" but I was fooled into thinking that the delta knobs must be something other than pots because they simply spun freely. That's kind of a fascinating implementation, though.

I was aware that the 2465 has a CPU that is polling the state of the front panel controls. I have been assuming that this is responsible for the apparent lag in the vertical position controls, but it occurs to me now that the position controls may be doing something more complex, e.g. some kind of velocity controlled range shifting?

I specifically bought a 2465 (no A, no B) for two reasons: first was that it appeared to have direct vertical and horizontal range controls, unlike later revisions that just had blank knobs and only displayed the ranges in the readouts. The second reason, of course, was to avoid getting a counterfeit 2465B that had been through the hands of a certain miscreant. It seems that I failed in the first goal, since there is much more fly-by-wire in the 2465 than I expected (not complaining: the 2465 is a wonderful instrument, it just takes some getting used coming from older devices).

-- Jeff Dutky


AngiesHusband
 

Ok - going after the two pots A6R3125 and A6R3150...
Pulled from the board and the back of them looks like this:
https://groups.io/g/TekScopes/photo/261895/0?p=Created,,,50,2,0,0

One has Philips screws holding the two halves together (#0-80 screws? 5/8" long?)
The other... has pins?
What is the preferred method for getting in here? Take a pair of flush cutters and cut the head off the pin, then push the body through the other side?

Paul


Chuck Harris <cfharris@...>
 

Hi Paul,

If I recall correctly, the pins are a plastic attachment pin, which has
a coarse thread. I think I recall that I grab the head with a pair of
sharp diagonal cutters, and pry. It will rotate CCW. Once the head
lifts a little, grab it with pliers and rotate CCW.

-Chuck Harris

AngiesHusband via groups.io wrote:

Ok - going after the two pots A6R3125 and A6R3150...
Pulled from the board and the back of them looks like this:
https://groups.io/g/TekScopes/photo/261895/0?p=Created,,,50,2,0,0

One has Philips screws holding the two halves together (#0-80 screws? 5/8" long?)
The other... has pins?
What is the preferred method for getting in here? Take a pair of flush cutters and cut the head off the pin, then push the body through the other side?

Paul