Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
Tektronix 2247A problem. Keep or return?
|
On Sun, Apr 6, 2025 at 01:09 PM, Jim Adney wrote:
Hmmm, I didn't know this was ever a possibility. Can you give some examples of=== I am sure there are others too but Agilent U1272A and HP3466A are two examples with AC only or AC+DC options. Ozan |
|
On 2025-04-06 4:09 PM, Jim Adney wrote:
Hmmm, I didn't know this was ever a possibility. Can you give someI can't recall a specific model number, but I know I've seen it. Keep in mind that "true RMS" is the square root of the sum of the squares of a series of instantaneous measurements. Mathematically, that is the DC voltage on a pure DC (i.e. constant value) voltage reading. You'd have to look at the mfr specs to know for sure. If there's a DC blocking capacitor (which you can always add externally, of course), there should be a spec for what amounts to a highpass filter. Steve Hendrix |
|
On Sun, Apr 6, 2025 at 10:31 AM, Steve Hendrix wrote:
Some meters give you only the ripple when on AC, but some also include the DCHmmm, I didn't know this was ever a possibility. Can you give some examples of DVMs that have this "feature", where there's no blocking capacitor on the AC input? This makes me wonder which reading, AC, DC, or both, would be the True RMS reading on a TRMS meter. I only have one TRMS meter and that's a Fluke 27/FM I bought from Walter at Sphere. I've probably never tried to use it to measure ripple; my default is my non-TRMS Fluke 23. I'll have to try the 27/FM next time. -- Jim Adney Madison, WI USA |
|
On Sat, Apr 5, 2025 at 8:21?AM Jay Czaja via groups.io <jmczaja=
[email protected]> wrote: This scope has me scratching my head. I had it running all night withoutYeah, until you've eliminated PSU problems, there's no point in troubleshooting anything else. I have a question. What is the best way to measure ripple on the LV supply. When I checked it the first time, I had the 2247A on an isolationI don't know that an isolation transformer helps, but maybe you want the two scopes plugged into the same outlet. It can't hurt to connect the ground clip close to the power rail. Tek always specifies PSU ripple at 20MHz bandwidth, so you want the 20MHz bandwidth limiter on. I don't think most 1X probes have 20MHz bandwidth, so you might want to just use flying leads or a coax with clip leads to hook up. Note that you can measure your noise pickup by turning the instrument under test off with your leads hooked up. |
|
On 2025-04-06 11:25 AM, Jim Adney wrote:
It may not be the BEST way, but the quick and simple way is to use the same handheld DVM that you use to measure the DC voltage.Be a little cautious with this. Some meters give you only the ripple when on AC, but some also include the DC measurement (true RMS includes DC). So if the reading doesn't change much when you switch from DC to AC, it's not telling you what you want to know. You could of course connect thru a capacitor; it takes only a little calculation to determine an appropriate cutoff frequency based on the meter's input impedance and the capacitor value. Steve Hendrix |
|
On Sat, Apr 5, 2025 at 07:21 AM, Jay Czaja wrote:
What is the best way to measure ripple on the LV supply.It may not be the BEST way, but the quick and simple way is to use the same handheld DVM that you use to measure the DC voltage. Just switch your DVM from DC to AC and you will get a very good idea of the ripple. In most cases, you're not looking for the difference between 2 mV and 5 mV; you're looking for the difference between 2 mv and 2 V. This is probably not useful for most switching supplies, since most DVMs don't go much above a few kHz on AC, but it has the advantage that you're already connected to the test points you're interested in, you automatically get a floating measurement, and it tells you whether you need to look into this more carefully or not. -- Jim Adney Madison, WI USA |
|
I misspoke. It looks that that particular .bin file can't be downloaded for some reason.
Luckily, 160-3493-00 / IC U2400 (I mistyped the p/n in my previous post) is socketed and there no need to desolder. I'm going to remove the IC and clean the IC pins and socket and see if that helps things any. I'll also try to swap the IC from my 2245A if cleaning it doesnt work. If that works, someone from the list has a parts 2246 with the needed IC. Thanks all! |
|
On Sat, Apr 5, 2025 at 05:21 AM, Jay Czaja wrote:
I have a question. What is the best way to measure ripple on the LV supply.No need to use an isolation transformer to float the scope for this measurement. The chassis of the scope is grounded normally, not like old CRT TVs or monitors that mostly had a "hot" chassis. Just use your other scope, set it to AC coupling and go as low in attenuation to get a readable signal across the power rail being tested. Connect the probe to ground and the tip to the power rail and see how much noise level you measure. You might also want to switch in the 20MHz bandwidth limiter so as to possibly get a more cleaner and relevant display. Ideally use a X10 probe, but X1 should be ok. |
|
On Fri, Apr 4, 2025 at 03:23 PM, Jay Czaja wrote:
I found also this part number 160-2393-00 on the following 2245A "TekWiki"pageYou could download that BIN file and program it into a new prom chip. Perhaps the existing one is flaky, hence the come and go error? |
|
This scope has me scratching my head. I had it running all night without any problem, I doubt part number 160-2393-00 is actually the culprit despite the occasional startup error. I suppose I should scrutinize the PSU more.
I have a question. What is the best way to measure ripple on the LV supply. When I checked it the first time, I had the 2247A on an isolation transformer with my 2245A connected directly to mains AC. I used a x1 probe on my 2245A to check for ripple and clipped the ground lead of the probe to the chassis of the 2247A. Would it be more accurate to connect the probe ground to the ground on the LV test header (J1204) instead of the chassis? |
|
I have a bit more info. I started the scope up again and I got the initial interrupt error that I posted in the beginning of this thread.
When I shut it off and started it back up, I was able to read the error message on the screen. It read "PART NUMBER 160-3493-00" I was able to find this in the service manual and it appears to be a microcontroller. "MICROCKT, DGTL:8 BIT MICROCOMPUTER, MASKED FOR READOUT SYSTEM" Is this a part I would need to find a donor scope to scavenge from? I found also this part number 160-2393-00 on the following 2245A "TekWiki"page and there appears to be a ROM image. Is this something that would be abled to be downloaded and programed? |
|
Alright, so I'm not in the clear with this 2247A. I've been busy with other things and had the scope off for about 3 days. Turned it on this morning and noticed the trace was a little noisy without a probe connected and I also noticed a very faint high pitch noise that I haven't heard before. I didn't think much of it but after being on for about 5 minutes, the scope locked up and all of the LEDs in the horizontal section were off and all controls were unresponsive. When I turned the scope off, I noticed the screen displayed a little square that disappears, almost as if something was being discharged. I really don't know how I can explain it but usually when I turn the scope off, the display just turns off. I noticed this screen aberration when turning the scope off when I initially observed the lockup/error problem.
When I turned it back on, the high pitched noise was no longer present and it hasn't locked up again. The high pitched noise would point to something in SMPS, or at least I would think. Any idea how I can troubleshoot this further? My though was to leave the scope off for another few days , take the chassis out before restarting it and then if I get it to fail, check the voltages and see if anything looks to be out of spec. It seems like the scope doesn't like a cold, cold start and the failure disappears after the scope stays on for a bit. Almost seems like one (or more) of the caps are borderline. Thoughts? |
|
Thanks again. I'll pull the processor board and have a look at the power supply and replace any diodes marked ZM or ZS.
I'll definitely replace the battery, I think Digikey has the correct one. I'll leave the electrolytics as-is unless I see anything that looks suspect. Had the scope off all night and cold started it in the morning without issues. I'll do this over the next fews days for good measure but thing are looking good at this point. |
|
On Sat, Mar 29, 2025 at 01:42 PM, Jay Czaja wrote:
Just check the power supply diodes for either ZM or ZS markings and replace. And the 3V battery. Should then be done and ready to enjoy. |
|
On Sat, Mar 29, 2025 at 09:57 AM, Jay Czaja wrote:
Thanks again guys! The 2247A has been running overnight with signals in eachJust replace the 3V battery to get rid of the annoying power up error message and the recalibration routine running every time. You need to lift the main processor board that is on top, and the battery is at one end of the board. To lift the board there are some flat ribbon type conductors and connectors to deal with, but if you are careful and methodical its pretty straight forward procedure. Last time I recall the battery swap can be done by just partially lifting the board, which limits the amount of ribbon cables to deal with. I have a really nice 2245A and enjoy using it and always wanted a 2247A.The 47A is a 45A with additional digital functions (and push buttons), otherwise they are identical scopes. Make sure to check those suspect diodes on the power supply board in both your 45A and 47A scopes. A quick glace should tell you if you need to replace them. The power supply board is in the middle of the lower side, kind of a pain to get it out and even more pain to get it back in due to the connecting pins. But you should be able to read the labels on a couple of diodes without removing the board to make sure you don't have any time bombs sitting in there. If you find voltages and ripple specs are ok I would not even bother recapping unless you see a capacitor that is obviously leaking or bulged up, which I doubt. These 224x scopes if I recall used high quality Mallory or Sprage caps, and I've never bothered to replace any in my 47A, but most definitively had to in my 2465B. Those old caps sometimes even end up measuring better specs than modern equivalent caps (value/ESR). |
|
I just cracked it open and checked the DC voltages and ripple. Everything was well within spec. The inside also looks very clean!
I didn't notice the tag earlier but it appears to have been taken from a hospital environment. It was a bit of a pain to pull the chassis from the case. Of course, the only tiny dent on the top of case was putting enough pressure on the side rails to make it difficult for the chassis to slide out without resistance. I'm guessing you might be right and maybe the caps needed to reform or something along those lines. At this point I had the scope running for hours. Left if unplugged for a few hours and had it running for a few more hours and everything seems to be ok. I'll probably recap it in the near future as I plan for this to be my everyday scope w/ my 2232 for 'slower' stuff. The 'green screen of death' had me a little worried but feel relieved that it's been stable since those initial issues. All functions I've tested so far have worked perfectly. I'll need to play around with the menu and measurements system a bit more but the few quick measurements I've made were good. I need to read about the menu/measurement section as it has more features than the 2245A which I've been using. I appreciate all the responses. I'm sure I'll have more questions! |
|
On Sat, Mar 29, 2025 at 12:57?PM Jay Czaja via groups.io <jmczaja=
[email protected]> wrote: Thanks again guys! The 2247A has been running overnight with signals inSounds like a keeper to me! If the scope has been in storage for a while, it's possible that the power supplies stabilize as the caps reform or warm up or some such. You may still want/need to recap it, and I've never heard of a case where the calibration is thrown off by improving the power supplies. I've never worked on a 2247A, though, so YMMV. |
|
Thanks again guys! The 2247A has been running overnight with signals in each channel and everything has been humming along ok, no lockups.
I'll put it on the bench sometime today or tomorrow and check the DC voltages per the service manual and check for ripple. I'll have a look at the diodes and check their marking and see what's what.. If any electrolytics require replacing, will this throw off the calibration? I don't have any of the fancy gear to do a proper calibrations (PG506A, marker generator, etc). I have a really nice 2245A and enjoy using it and always wanted a 2247A. This particular one seems to be clean and in very good condition overall. The trigger section on the 2247A seems better than my 2245A as well. The 2400 series scopes look nice but the 224x's are probably more scope than I'll ever need. I also have a nice 2232 that I like using for the DSO section. I mess around with building synthesizers and the 2232 is nice for catching really slow events. Thanks again! |
|
On Fri, Mar 28, 2025 at 06:12 PM, VE6WMR wrote:
Jay,Comparing a 2465 to a 2247A is a bit like apples and oranges, but understood the well meaning intention to be helpful in the sense that both are old. But for one luckily the 2247A does not rely on NVRAM to store cal constants. As suggested check all voltages and ripple to make sure they are within spec. Also while you are in there, check power supply diodes, and any with ZM or ZS markings needs to be replaced as they are known to get leaky over time and cause problems. There are 17 of these diodes on the 224x (and 2252 same series) scopes power supply. Use fast diodes like MUR160 (600V 1A) or similar. Do not use common 1N400x. Also replace the 3V Lithium battery, its only meant to hold the setting of the last session, no calibration at risk. I have both 2465B and 2247A scopes, but the 47A is usually my go-to scope 95%. |
Jean-Paul
to navigate to use esc to dismiss