EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: GerryR on October 05, 2019, 05:49:48 pm
-
I am getting a "Test 04 Fail 03" diagnostic error on power-up, which I understand is a calibration related error, but searching online and through the manual, I have been unable to come up with the details of the "03" part of the error. Any help on specifics other than the general "Calibration Related Error?" Thanks in advance.
-
03 = Parity error with out of limits and good checksum
When the CMOS RAM's battery goes dies, or there a problem with the CMOS ram's parity, the scope may detect that as an error, and the Test 04 Fail XX message will be displayed.
-
03 = Parity error with out of limits and good checksum
When the CMOS RAM's battery goes dies, or there a problem with the CMOS ram's parity, the scope may detect that as an error, and the Test 04 Fail XX message will be displayed.
How is this remedied? There were no errors before the battery was changed (battery was dated 07 87!), so I am assuming that calibration data was corrupted, and a re-cal is in order?? Am I on the right track (unfortunately)?
-
TEST 04 Error .... Flow charted
https://youtu.be/YubzXhE27dY
There are TWO Batteries in tbe 2400 series scopes.
Did you change BOTH ?
-
I've been through the "Flow Chart" for Test 04 failure; unfortunately it doesn't address the "03" error. I only replaced the battery on the A5 control board. I didn't know there was a second battery. Can you enlighten me as to where it is? Also, in the video, he mentions "hidden" calibration procedures. Is there a list of these somewhere? Thank you for your input.
I did check the cal ram, and from 00 to 6E there should be odd parity. About 10 of the last locations have even parity, which I am assuming is the problem. I haven't gotten far enough to figure what that means and what parameters are in those locations.
-
03 = Parity error with out of limits and good checksum
When the CMOS RAM's battery goes dies, or there a problem with the CMOS ram's parity, the scope may detect that as an error, and the Test 04 Fail XX message will be displayed.
How is this remedied? There were no errors before the battery was changed (battery was dated 07 87!), so I am assuming that calibration data was corrupted, and a re-cal is in order?? Am I on the right track (unfortunately)?
Your assumptions are correct and you are on the right track. A re-calibration is in order. This will fix the errors that were flagged when you changed the battery.
-
...
There are TWO Batteries in tbe 2400 series scopes.
Did you change BOTH ?
There's only one battery in the 2465A.
-
...
There are TWO Batteries in tbe 2400 series scopes.
Did you change BOTH ?
There's only one battery in the 2465A.
And I believe no battery in the 2465.
-
...
There are TWO Batteries in tbe 2400 series scopes.
Did you change BOTH ?
There's only one battery in the 2465A.
And I believe no battery in the 2465.
Correct. Uses EAROM to store cal data.
-
...
There are TWO Batteries in tbe 2400 series scopes.
Did you change BOTH ?
There's only one battery in the 2465A.
I would call that negative development progress, where a dead battery makes a 'scope almost useless without recal. Maybe they didn't expect people to hang onto them that long!
And I believe no battery in the 2465.
Correct. Uses EAROM to store cal data.
-
...
There are TWO Batteries in tbe 2400 series scopes.
Did you change BOTH ?
There's only one battery in the 2465A.
I would call that negative development progress, where a dead battery makes a 'scope almost useless without recal. Maybe they didn't expect people to hang onto them that long!
And I believe no battery in the 2465.
Based on the OP's posts, it appears that the scope did not have a "dead battery" rendering the scope useless. Unfortunately, the loss of calibration information appears to be a self-inflicted wound. On the bright side, the unit can have a fresh makeover on the passives that typically fail on this unit, cleaning, etc. prior to performing the calibration procedure. All is not lost. Just an opportunity to keep this scope running for many years to come.
-
You are right, the battery was not dead, but the date code on it was 07-87 so how much longer would it have gone - who knows. However, a word of caution. I thought I took every precaution when installing the new battery. The board was out of the 'scope, obviously, and a substitute battery was attached before removing the old battery. Sounds good, so far. No problem installing the new battery, but Test 04 Fail on power-up. So what went wrong, and how did the cal memory get contaminated? Just a guess, but I was working on a static free mat, which I had laid the board on. I had problems years ago working on a static free mat with cmos circuitry; I didn't give it a thought, but there is enough conduction in the mat to screw things up. Just a guess, and a word of caution.
-
You are right, the battery was not dead, but the date code on it was 07-87 so how much longer would it have gone - who knows. However, a word of caution. I thought I took every precaution when installing the new battery. The board was out of the 'scope, obviously, and a substitute battery was attached before removing the old battery. Sounds good, so far. No problem installing the new battery, but Test 04 Fail on power-up. So what went wrong, and how did the cal memory get contaminated? Just a guess, but I was working on a static free mat, which I had laid the board on. I had problems years ago working on a static free mat with cmos circuitry; I didn't give it a thought, but there is enough conduction in the mat to screw things up. Just a guess, and a word of caution.
You can try one more thing but I'm not sure it will help. Try the EXER 03 (Clear Cycle Error) routine listed in the service manual. If that doesn't clear the error, then it's definitely calibration time.
-
Unfortunately, been there, done that; it needs the cal. I was wondering if anyone here has done the cal, and if there is a minimum set of procedures that will get it back up and running. The procedures in the manual are not particularly complicated, but some of the required test equipment are not the usual home-lab stock items. Does anyone know of any work-around procedures? I have a few ideas that I am going to try before I think about sending it out; if they work, I will post the results.
-
Gerry the 2465A cal takes about 3-6 hrs with exactly the specified TEK cal gens, if one is experienced.
Many steps are very critical in gen requirements.
Failing ( out of limit) one step, results in no CAL data save.
See the very fine 2465 teardown threads (63 pages) and the TEK scopes groups.io for more information.
I have worked on 5 of them now and still get some steps wrong!
But lots of fun to learn the proceedure.
Enjoy
Jon
-
In case of scrambled calibration constants only (due to dead battery for example) the calibration will not be that long.
All you have to do is basically 2 or 3 sections out of calibration routines, I believe CAL-01, 02 are a must Maybe 03 or more, don't remember exactly, but it is easy to figure it out as you go along.
Of all these only CAL-01 (horizontal) is rather involved and requires rather exotic pulse gen. capable of 2ns pulses, tek TG501 is the one that tektronix requires in the manual.
They call it a marker generator and I really suggest you get a hold of one.
The second thing you'll need is a "Calibration generator" , again tek calls for PG506,
but this time is not strictly necessary.
Even though I do have one and used it to calibrate 2465A, I repeated the calibration using modified FY6600 and it went fine.
Even doing it for the first time it didn't take me more that 2hr total, with practice it should take much less time.
Couple of pointers, make sure that you're following the CAL procedure for the correct serial number of your scope ! (cal. procedures for later serials are at the very end of the manual).
You will need another scope for the procedure, will be easier if that scope bandwith was >100MHz (I've done it with 1054Z, but at the fastest sweeps in CAL-01 it was sort of a guesswork :-[), also you will need a 50 \$\Omega\$ inline terminator if your second scope doesn't have 50 \$\Omega\$ inputs.
A couple of BNC cables and you god to go.
And don't be scared it is fun (you will learn how to count to 28 :) :)).
-
Alpher, the cal routine for lost memory really should have every step and start with power supply and DAC.
Proceedure depends on serial number, options and model number.
The transient align is difficult as it needs some 50 ohm precision cables and 2, 2.5 and 5X BNC attenuators and very fast pulse generator, like PG506 or Leo Bodnar pulser.
Doing a partial cal is OK but the scopes will not be in spec.
Finally some of the steps interact and are itérative.
Just my thoughts
Jon
-
I'm too old to be scared of much in electronics; just lacking some equipment and accessories for this calibration. I have been through some of the calibration procedures; messed up some of the Cal 01 procedure due to a lack of understanding of some of the requirements. I downloaded the PG506 and TG501 manuals to get some clarification of what signals were needed for the calibration. Just waiting for a couple of attenuators that I ordered. I did make a "Normalizer," as it seems that anything related to these older 'scopes is thought to be "gold." And, I will detail the test procedure that I use, and the equipment / accessories, just in case I live long enough for the new battery to die. :-\
Thank you for all the input.
Just for the record, I contacted Tektronix for a calibration quote and was told they no longer (since 1993 ::)) "support" the 24XX series scopes. I would think that "calibration" would fall under a different category, but what do I know. Also, just for reference, I have quotes in the $350.00 to $500.00 range from independent cal companies for the 2465A. I can get some pretty decent equipment to do the job for that amount.
-
Alpher, the cal routine for lost memory really should have every step and start with power supply and DAC.
Proceedure depends on serial number, options and model number.
The transient align is difficult as it needs some 50 ohm precision cables and 2, 2.5 and 5X BNC attenuators and very fast pulse generator, like PG506 or Leo Bodnar pulser.
Doing a partial cal is OK but the scopes will not be in spec.
Finally some of the steps interact and are itérative.
Just my thoughts
Jon
Does the transistent align stores anything into the calibration constants memory?
I really don't remember.
-
The transient align cal has some steps that are L, C or trimpot adjustable but other steps are indeed stored in the cal constant memory.
These interact and can be trimmed iteratively.
Contact me via email if you want a quote from a pro ( not me !)
Jon
-
I started into the recal (still waiting on a couple of components), but I have a couple of questions. Once finished with the partial calibration, I got a Test 04 fail 02, which is an "out of limits" failure. Is there a way to tell what is out of limits? Also, can anyone tell me what the pulse looks like coming out of the Tunnel Diode Pulser? From the schematic, it looks as though transistor turns on and when the threshold of the tunnel diode is reached, a sharp falling edge is produced. Is this correct?
-
You don't need tunnel pulser for restoring scrambled calibration constants.
-
You don't need tunnel pulser for restoring scrambled calibration constants.
Why not?
-
Strongly recommend to read the full calibration proceedure.
The transient response calibration uses a fast pulse eg PG506.
The very fine Leo Bodnar pulser is only £50 and is 40 ps edge.
The cal constant memory indeed holds some transient cal data, but many adjustments are L, C or R.
Just the ramblings of an old retired EE
Enjoy
Jon
-
Strongly recommend to read the full calibration proceedure.
The transient response calibration uses a fast pulse eg PG506.
The very fine Leo Bodnar pulser is only £50 and is 40 ps edge.
The cal constant memory indeed holds some transient cal data, but many adjustments are L, C or R.
Just the ramblings of an old retired EE
Enjoy
Jon
I like the ramblings of old retired EE's; I are one too! ;)
-
Another question or two for those in the know: At the beginning of the Cal 02 procedure, it states to connect a 0.5 V standard-amplitude signal to CH 1 and to use CH 2 Position to vertically position the "trace" to within 1 division of the center graticule line, and then to use the CH 1 Position and Volts/DIV VAR controls to obtain a 10-division horizontal signal. I don't get a "trace," only 2 dots. I can position the dots with the controls to put one dot on the first graticule line and the other on the 10th.
Question 1: "Is this "normal," or am I doing something wrong?" I've read and re-read the procedures and can't find any references to this.
Also, at the beginning of the Cal 01 procedure, it states that, "Upon entering Cal 01, the Input Coupling is automatically set to 50 \$\Omega\$ DC ....." This doesn't appear to be true for the other Cal procedures (2,3,4,etc) as I had to install a terminator on my generator to get the actual levels called out in the test procedures (monitored on another scope).
Question 2: "Is this correct, or am I missing something here, as well?
Thanks in advance for any input!
Gerry
-
Gerry many thanks for the note
Been thru the CAL proceedure a few times, great learning curve and many details and pitfalls.
On groups.io TEKSCOPES much more about this, especially TEK pros like Chuck H.
On eevblog see the 65 page 2465B teardown thread.
About the out of limit message, no further information is provided and usually its the generator signal at fault or proceedure is not precisely followed.
Just re-read and do that step over.....and over....eventually one gets it!
Bon courage et bon chance
Jon
-
Gerry just saw the 2 new questions
Question 1/ Trace/ dots....my recollection is the dots are the "trace" but may depend on intensity setting.
Question 2/ CAL 01 Term.....
Normally the 246xB INTERNAL 50 OHM term setting is used. Is is set automatically in some steps.
Any external termination will not be as good a VSWR or transient response.
I never saw that problem.
Enjoy
Jon
-
A little update. I was having problems with CH 2 during calibration (CAL 02). When I got to step 130, I would get a limit error. In normal mode, CH 2 relay would not switch when going from the 500 mV to 1 V scale. I checked out the logic to make sure the shift registers and relay drivers were functioning, which they were, and then proceeded to remove the attenuator. I checked out the individual relays and found one that would not transition from one state to the other. I dismantled the relay and found burrs on the sides of the "rocker-contact." I removed the burrs with a fine diamond file, re-installed and tested; relay was then fully functional. I re-installed the attenuator into the scope, powered it up, and now the relay changes state when going from 500 mV to 1 V. I haven't rerun CAL 02 to see if that problem went away with the relay fix; tomorrow is another day. Pictures attached if anyone is interested.
The last photo shows part of the ceramic substrate of the attenuator with some "dark areas" on the left side. They look like traces of carbon (though they don't rub off) but I'm not familiar enough with these attenuators to know if this is normal. Any input is welcomed.
-
Wow, you really ventured in deep. Good fix! :-+
Dunno about those dark areas on the substrate. It almost looks like arc tracks to me but that makes no sense because the attenuator would blown sky high. :-DD
-
That's what I thought at first, but nothing rubs off like you would expect from a carbon track. Well, testing will tell all.
-
Gerry kudos for the attenuator relay job, and fine macrophotographie.
The "carbon" areas could be the normal laser trimming of thick film attenuator resistance.
Notice how only certain resistors are affected and how deterministic the carbon areas are?
But never opened one up so this is just speculation.
Perhaps one of the TEK mavens or retired engineers can comment?
Just the ramblings of a retired EE
Jon
-
Believe it or not, my left hand was holding a 5X eye loop over the substrate, while my camera was in my right hand taking the picture; nothing fancy!
-
Gerry looking again at the last photo, I am now sure the blackened areas are from trimming of the thick film attenuator resistors, and not due to overload.
Jon
-
I believe you are correct. I powered up the 'scope and the attenuator is responding correctly on every scale. Now back to the re-calibration. :)
-
Well, that CH 2 relay was what preventing full calibration of the 'scope. The dreaded out-of-cal ???????? are gone. Picture below. I now know more about the 2465A internals than I ever wanted to know, all thanks to some mishap during a battery change. Thank you all for all your support!
-
Gerry: GOOD JOB, especially on the relay repair and learning the CAL procedure.
If you look on the 2465 teardown thread, and groups.io, you will see many notes for backup of the NVRAM, battery replacement, etc.
Even after CAl, I would recheck H and V cal and the measurement routines as a double check.
Easy to miss the transient and CRT CAL.
Enjoy,
Jon
-
I have been through the teardown thread and on groups.io. In fact, I followed the procedure for changing out the battery, but something went wrong, and as I stated earlier in this thread, I believe laying the control board down on an anti-static mat may have done me in. I had a problem with that several years back and cooked a cmos circuit while working on a similar mat when I powered it up. It's fine when working on non-powered boards, but the battery to the ram slipped my mind. Oh well, I'm back up and running and learned a lot along the way.
I really intend to go through the full cal procedure again, but want to use the 'scope for awhile to get used to its operation and potential. I have a circuit that I need to build that will utilize this 'scopes bandwidth, but one of these days (I sound like Pa Kettle!) I'll be back into the calibration, again.