Tektronix Technical Forums are maintained by community involvement. Feel free to post questions or respond to questions by other members. Should you require a time-sensitive answer, please contact your local Tektronix support center here.
I have a TDS510A with diagnostic log error (see also photo):
acqMemAddr **addr=0x7300000 exp data =0x101 actual....
acqmemdata **addr=0x7320000 exp data =0x5555 actual....
vertAttenShiftReg **exp data=0 actual=1 for bit #31....
knowing that I've changed all the capacitors, verified all the traces below these capacitors (all were OK), verified that all capacitor had good voltage across, and read identical forum without founding a solution, here are my questions:
How can I do to see the whole message on the screen after actual.....?
Is it a problem of memory? As I understood on several discussion it could be U201 of multiplexer A but not sure...on other forum that was not the solution...
Could somebody help me to find the problem?
- 20181226_214942_resized.jpg (355.47 KiB) Viewed 3876 times
Hmm, I forget if you can see the full messages from the scope's UI. I don't think the extra details would help you much, though. Does look a bit like bad memory, but as you say it could be a bad mux too. If you really want to see the full errors, one way is to connect to the internal "service console port" connector.
Yes I finally realized a debug link where I've found the complete listing. I've attached this listing on the message. I can see message like :
** exp data= 0x1 actual= 0xfdb7 for bit # 0 for addr: 0x72c2000
Since I've connected this debug link, the scope is hanging at the same state:
0x51f0018 (V main): duart portB write timeout, byte: a1
and no more is displayed....
The 0x72C2000 address is the DSP/DMUX CH1.
maybe one mux is dead but difficult to know which one....
- board listing.txt
- (9.24 KiB) Downloaded 137 times
just had a quick look, and I see these
Code: Select all
ERRORID: 62 hardware error acq processor not responding ... -> ERRORID: 64 wrong hardware configuration Dsp D1 mem siz probe, expect: ffff, actual: 0 ERRORID: 64 wrong hardware configuration Acq number of digitizer probe, expect: 4, actual: 0 ... acqProcThermistor...............ERRORID: 62 hardware error acq processor not responding
Yes! you're right, the board was not connected in the correct sense. Thank you very much!
Now I come back to my first problem (first mail I've made on this topic), I put the listing in copy.
I can read the following text:
ERRORID: 150 diagnostic test failure acqMemAddr
** addr = 0x7300000 exp data = 0x101 actual = 0x3939
*** WAS TESTING DIG A ***
ERRORID: 150 diagnostic test failure acqmemdata
** addr = 0x7320000 exp data = 0x5555 actual = 0xaafe
*** WAS TESTING DIG A ***
I suspect the multiplexer U200 to be the problem because U200 is cold (or not so hot than the others) and the other U100-300-400 are hot when the scope is operating.
Maybe U200 is in stand by mode because of bad test on associated RAM? Or U200 is defective?
How can I be sure that U200 is operating properly?
Thanks for your help,
- listing 190219.txt
- (5.91 KiB) Downloaded 122 times
Interesting observation. I'm not sure what to suggest. It's possible to call some of those functions (acqMemAddr tests, etc) from the service console, it might be possible to get them to test other areas or other muxes specifically. To my knowledge none of this is documented anywhere; it needs to be figured out by analyzing the firmware dump...
I've found the problem !
It was the clock of multiplexer U200 !
Signals DCLKT and DCLKF had not the same shape than the others mux.
These signals came from U750. On this chips, the inputs are CLKC and ~CLKC which had bad levels. These signals cames from U502. The input links to these signals is ADLY on U502 which had not a level between -3 to -4,5V as written on the schematics but was at -10.2V.
The origin of the problem was the U506D amplifier from a TL074C.
Changing this chip was the solution, now my scope displays waveforms!!!
I don't see any problems but on the debug link I can see :
ERRORID: 150 diagnostic test failure ctlConfidencDiag
** SMUT was NOT asserted
** SDUT was NOT asserted
** 2.50e-9 <= exp <= 1.20e-8 actual= 0.00e-1 main TI failed
Where does it come from?
I would like to thanks Mr Fenugrec and Peter Strick for their help !
The TI stands for Time Interpolator. It's part of the trigger circuit and has lots of SMCs around to leak and cause problems. I've spent so much time fixing things in this area that I made a full series of waveform readings on my schematic. Right now it's on a 13x19in sheet, but I can copy parts and will post it later. Virtually any deviation from the waveforms seen here is cause for further troubleshooting.
BTW, the diagrams and trouble shooting logic come from the TDS-520 Service Manual, Vol II. It helps with any of the 500 and 500A series. I believe the only legit source is Artek Media. I have found this invaluable in fixing these old guys...
sorry about the "K" last time.
attached are the waveform and voltage readings from the TI1 and TI2 circuits. They are pretty much the same. The "RAMP" (pin3 of U1603 and U1653) is discussed in the service manual (page 9-25) as varying from 6 to 20 Usec. This signal is complex. When you look at the signal (with another scope, of course...), at timing of 10 ms/div, you will see a negative going pulse with a width of about 10 msec, with a positive spike on the leading edge. When you decrease the time scale towards 10 Usec and move the trigger point to positive voltage, then you will see the ramp moving from zero to +1 volt and sweep down to -1 volt in the 6 to 20 Usec.
If the main trace prior to the pulse is not zero, then you have a problem in the upper or lower circuitry that feeds it, or in U1602/U1652 (for TI2).
the scans are three pages each. Note that the voltages for TI2 are shown and are the same for TI1. Also, TI2 has the ramp when not triggered, but I believe TI1 needs it triggered or the ramp will change to the waveform shown in the lower diagram.
Thanks for your help.
I spent a lot of time to verify the time main interpolator. I suppose that this main interpolator is on the sheet "time interpolator and jitter ramp 1" that is sheet 13 of A10 of TDS544 schematics equivalent to TDS510A.
So I've verified the signals and all seems to work well, when scope triggers, I obtain the signal tek40 (yellow is RAMPSW1 or TP1603, red is pin 3 U1603, green is pin 7 of U1603) , voltage decreasing slope last 24us and rising slope (tek 41) is 8,5ns so well between 2,5ns and 12ns ( the default is 2.50e-9 <= exp <= 1.20e-8 actual= 0.00e-1 main TI failed).
When the scope is not trigged, the curves on tek42 appears, it seems to be also OK.
Comparing these curves with working TDS510A give also the same waveforms.
The scope appears to trig well (maybe a little sensitivity in DC coupling, sometimes scope lose the trig and then I've to change vertical sensitivity and come back to initial vertical sensitivity to get trig again but I'm not sure that this is the problem).
So to conclude, I would say that the scope is working well but I still have this default:
SMUT was NOT asserted
SDUT was NOT asserted
2.50e-9 <= exp <= 1.20e-8 actual= 0.00e-1 main TI failed.
What are SMUT and SDUT signals?
How is it possible for this scope to measure such tiny times?
Any idea to get rid of this default?
Maybe I've to set the scope to factory default settings ?
Thanks for your help!
- tek00042.png (24.94 KiB) Viewed 2337 times
- tek00041.png (24.66 KiB) Viewed 2337 times
- tek00040.png (27.05 KiB) Viewed 2337 times
The waveforms look correct to me. SMUT and SDUT come from U1001, the main ACQ controller (the guy with the pancake cooler on top). Real unlikely that he has a problem. I've changed these and it never helped. And it has lots of pins to desolder. I would still suspect an error in the TI circuits or things leading to them.
1. If TI1 and TI2 both look good, then rerun self test after the scope has been on for a few minutes. My 520A test scope often throws this a TI1 error (but not the SMUT) on initial boot, but clears in a few minutes. TI1 always looks good.
2. Go back to your original solution at U506. Make sure something else in the immediate area has not been corrupted also.
3. Check the acquisition clocks on U501 in addition to 502.
The losing the trigger that you mention in DC should not happen. You might check the voltage levels from the Analog DAC control and ensure that every single signal that is shown as an output is a steady voltage that varies only with the control that owns it. E.G. Offset 1 (pin 4 of U932) varies only with changing channel 1 offset.
Yesterday I began, as you suggested me, to come back to U506 and I measure different voltage between my TDS510A and the operating one from my friend. (the scopes are in the same conditions, no input on CH2 CH3 CH4, only 1kHz, 0,5V signal on CH1, same horizontal and vertical sensitivity)
On the output of U506D pin 14 I measure -4,555V and on the output of U506A pin 1 the voltage is 1,636V, voltage seems to be stucked at these values all the time. Calculating gains on U506D it seems to work well but the voltage are differents on the working scope on which I measured on pin 14 -2,63V and at pin 1 -2,2V and these voltages varies slightly with time.
If I go deeper in the schematic, I saw the COLY signal is different on my scope and the other and it's too difficult to analyze because on U941, COLY signal comes from multiplexed +/-1,71V signal....
Moreover I measured a different clock frequency on U501 and U502, on my scope ADCLK frequency is around 3,4KHz but on the working one frequency is 1,5KHz.
Are these differences normal?
I also tried to start again the scope after long duration of operation but default of the main TI is always there....
U501 and U502 have ACQCLK inputs that should be 250 mhz and very close in magnitude on pins 15 and 16. Pins 2, 3, 27 and 28 are outputs and should all be 250MHZ also. I'm not sure what ADCLK is.
I did replace one of these when it had the proper inputs, but one of the outputs was zero.
Did you get the TDS520 manual? It has the only trouble shooting guide for the 500 and 500A series scopes that I know of. You can get it from Artek Media for a reasonable price and he emails you a link to download.
Sorry for my late answer, I was waiting for a scope able to measure 250MHz....
For the schematics, I use those of the TDS540 and they are very close to the 510A.
For the clocks, I measured on pin 15 and 16 of U501 and U502, and they are the same. I've compared pin 15 of U502 of the working scope (red) and mine (green) and the signals are very close see -.
All the outputs on pins 2, 3, 27 and 28 are at 250MHz, I've also compared pin 2 of U502 of the two scopes.
The only differences is at pin 6 ADLY of U502, on my scope (green) the voltage is stucked -4,5V and on the working scope (red) it's around -1,4V but it's varying slightly with time, see -. Difficult for me to know why these voltage are different because it's come from a multiplexed signal on U941 (signal COLY is demultiplexed from +/-1,71V signal).
I'm not sure these differences of voltage is the origin of my "2.50e-9 <= exp <= 1.20e-8 actual= 0.00e-1 main TI failed" default.
Maybe the scope must be recalibrated to avoid this problem?
What do you suggest?
Is there anybody who had the same problem?
Thanks a lot!
- pin 6 of U502.png (16.48 KiB) Viewed 1259 times
- pins 2 U502.png (21.22 KiB) Viewed 1259 times
- pin15 U502.png (18.72 KiB) Viewed 1259 times
My voltages for pins 6 and 7 (ADLY and BDLY) on both U501 and U502 are -3.0 volts. The signal at pins 3 and 5 of U505 and U506 are both about -1.65 volts. If your voltage at pin 3 of U506 is in the neighborhood of -1.65, then you have a problem in U506. I assume that if this voltage is greater, then some kind of delay is being asserted and causing the lower voltage. I assume an increased voltage at U506 will cause ADLY to decrease as the range is stated as -3 to -4.5 volts.
hope this helps.
My voltage at pin 3 of U506 is +1,63V. The voltage at the output of the amplifier is Vout=Vref+(10K/22.1K)*(Vref-Vin)=-2,618+(10/22.1)*(-2,618+1,63)=-4,54V and I measure -4,55V so I think U506 is OK.
But I've remarked that pin2 of U502 (=+CLKA) and pin 28 of U502(=+CLKB) and not in phase on my scope but they are in phase on the working scope see -.
I've tried to decrease R540 and decrease the gain to decrease the voltage at pin 14 of U506, I've obtained -3V with 1K in parallel with R540, restarting the scope always the same default....
I not sure if these two clocks must be in phase, if so, I think the system try to compensate a delay but not succeed, but where this delay comes from?
The signal at pin 3 of U506 comes from multiplexed signal named +/-1,71V....difficult to know more about it...
Thanks a lot!
- pin28green_2red_myscope.png (24.65 KiB) Viewed 1119 times
- pin28green_2red_workingscope.png (25.18 KiB) Viewed 1119 times
Pins 2 and 28 of U501 and 502 seem to be in sync on my 544A. However, I noticed that during boot up, it looked like both pins2 of 501 and 502 were delayed relative to pins 28. However, they synced up after boot up. See the attached screen capture. Actually, when I only was measuring 502 the first time, they showed a big phase difference (2.5 ns out of 4 ns), but it cleared when I hooked up all four. Thus, a phase difference may be simply related to some setting.
You indicated earlier that the scope seemed to work fine, except for the error message. If it really does work fine, then you might just ignore it or try to dig deeper to find out what doesn't work like it's supposed to. That way we have a way forward. The phase and voltage difference that you are seeing might be just fine or it might not...
At this point, I would normally try to calibrate the scope and see if the calibration routine found something it didn't like about the scope. If you can't do that, you might do the same thing manually by running all the performance checks in the 544A manual. It's easily available -. If you don't have the right stuff to do it exactly, you can improvise with whatever you have to drive the signals. That way you might find something (triggers, amplitudes, whatever) that isn't right.
- u501-502.png (12.04 KiB) Viewed 1110 times
I retrieved the performance check of the TDS540.
I will try to do follow some verification procedure, I began to recall factory setup without success...
One more question:
I use the square wave test signal for probe compensation on CH1. The scope trigs correctly the signal. But if I move the level trigger outside the signal the scope does not trig the signal anymore, this is normal, but if I put again the trigger level in the range of the signal, the scope does not trig the signal anymore...To trig again, I have to change vertical sensitivity or horizontal sensitivity and come back to the same sensitivity and then scope trig again....
If I put the trigger in HF reject no more problem....the trig is OK even if I put the level outside the signal and come back again. The problem happens only in DC coupling and I think it is not due to noise because before moving the trig level the scope trigs correctly.
This is not normal and as it is a trigger problem maybe this is link to my 2.50e-9 <= exp <= 1.20e-8 actual= 0.00e-1 main TI failed" default?
Does this problem suggest you any idea?
That really sounds like a TI1 problem. Let me duplicate what is causing your trigger problem and I'll send directions and what the results should be. Of course, you will need another scope to examine the signal at a couple of points to do the job.
Who is online
Users browsing this forum: No registered users and 5 guests