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
465B Horizontal Trace Issues - Help requested
|
I have a 465B/DM44 that is new to me that I think has issues. I am a relative newby to scopes, but this one does not seem right. There is "compression" of the horizontal trace, but it varies. Sometimes the trace just shrinks to a width of about 1-1.2 major graticule (1/2" in the same spot), other times it only shrinks a bit, covering about 7.3 graticules, with the best it ever gets being about 9.6 (full screen = 10). This significant compression happens randomly, both when scope is relatively "cold' and after it is fully warmed up. It affects both channels. I've used different probes - no difference. I can impact this, by moving the sweep left to right with the horizontal position control. This sometimes moving the horiz position "fixes" it, other times not. If it is temporarily working OK, I can get it to shrink if I aggressively manipulate the horiz position knob. In addition the "timing" may not be correct - not sure I am saying this correctly, but the calibration square wave gets "smaller" horizontally as the trace moves from left to right. I would expect the length of the square wave to be the same all across the screen. I suspect this is related to the compression, but not sure - seems relevant to mention. The only thing I have done is removed the cover and made sure the connections to the deflection plates on the CRT were secure (they appeared to be). Everything looks clean and untouched inside, and overall the unit is clean and does not look abused. I have not cleaned any switches or controls, but that seems like a good place to start. If it matters, serial number is above 60xxx, which I believe makes it the later series. I have downloaded the manual from the TEK site, but they only have the low S/N version. If it would help I have photos and even short video of the bad behavior.
I come here seeking help, on how to proceed, possible causes, how to troubleshoot. Thanks in advance Gary. |
|
Hi Gary:
See, to download Tektronix's superb and easy to follow (compared to the stuff we get today.) There is a lot of info on the Web about this very popular scope... just Google. I suggest, read the 465B user manual to make sure you understand what the unit can, is supposed, to do. (Based on that understanding, you might have a better understanding of what issues (if any) the unit has. A search on messages in this forum will also yield plenty of info/tips on these scopes. Yes, post a link to your photos/video on Google Photos, Imgur... and video to YouTube... and people on the forum will take a look. There are many very knowledgeable people on this forum, who generously share their time. Let us make best use of that. Regards. Roy |
|
I have faced a similar problem on my 465B. In my case, horizontal sweep used to shrink and expand at random. There was no linearity issue, as calibrator square wave looked uniform across the width of display, only shrunk randomly to anywhere from six to ten horizontal divisions. Fault was traced to an intermittent x1 gain setting potentiometer in the horizontal amplifier. This pot was replaced and gain adjusted as per service procedure, and scope has worked fine since then.
In your case, it may be a noisy horizontal position potentiometer, in addition to x1 gain potentiometer. Since you have linearity problems also, as you see shrinking of square waves at some positions of screen, there could be linearity problems in sweep area, or some limiting due to a failing component in one half of the horizontal amplifier section. My two cents to share, hoping you can get some ideas to explore. Shailendra |
|
Hi Gary,
The very first thing you need to do is to measure the low voltage test points for DC level and ripple. They are marked on the board. These scopes usually have bad electrolytic capacitors and you need to test for that. The service manual will tell you what to measure and what to expect. The correct manual can be had here. Regards, Tom On 9/20/2020 7:41 PM, gary@... wrote: I have a 465B/DM44 that is new to me that I think has issues. I am a relative newby to scopes, but this one does not seem right. There is "compression" of the horizontal trace, but it varies. Sometimes the trace just shrinks to a width of about 1-1.2 major graticule (1/2" in the same spot), other times it only shrinks a bit, covering about 7.3 graticules, with the best it ever gets being about 9.6 (full screen = 10). This significant compression happens randomly, both when scope is relatively "cold' and after it is fully warmed up. It affects both channels. I've used different probes - no difference. I can impact this, by moving the sweep left to right with the horizontal position control. This sometimes moving the horiz position "fixes" it, other times not. If it is temporarily working OK, I can get it to shrink if I aggressively manipulate the horiz position knob. In addition the "timing" may not be correct - not sure I am saying this correctly, but the calibration square wave gets "smaller" horizontally as the trace moves from left to right. I would expect the length of the square wave to be the same all across the screen. I suspect this is related to the compression, but not sure - seems relevant to mention. The only thing I have done is removed the cover and made sure the connections to the deflection plates on the CRT were secure (they appeared to be). Everything looks clean and untouched inside, and overall the unit is clean and does not look abused. I have not cleaned any switches or controls, but that seems like a good place to start. If it matters, serial number is above 60xxx, which I believe makes it the later series. I have downloaded the manual from the TEK site, but they only have the low S/N version. If it would help I have photos and even short video of the bad behavior. -- This email has been checked for viruses by Avast antivirus software. |
|
Tom,
I have the correct manual and measured DC voltages with 115.0 VAC line input through my Variac. Here are the DC voltages I measured: TP 4338 +55V, tolerance +54.62 to +55.39 Reading: +56.94 TP 4339 +15V, tolerance +14.75 to -15.26 Reading: +15.49 TP 4337 +5V, tolerance +4.92 to +5.09 Reading: +5.172 TP 4439 -8V, tolerance -7.86 to -8.14 Reading: -8.27 TP 4340 +110V tolerance +106.7 to +113.3 Reading: +113.5 It looks like all the readings are just outside of the acceptable range. I took the readings several times over a few minute span, and they were the same. I'm not sure how to measure ripple without another scope. Is there a way with a DMM (Fluke 117)? I'm assuming I took the measurements in the right place. I was expecting the TPs to be marked with the TP# (e.g. TP4338), but they were only marked with the voltage. These were all together on the main board on the bottom of the unit. I also shot some Deoxit into the position switch to see if that would help, but it did not seem to make a difference. What next? |
|
All the supplies reference to the +55 supply so you will want to adjust that first. Go to the adjustments section of the service manual (section 4-23) and follow the process. If you don't have a HV probe, you could probably just leave it as found.
toggle quoted message
Show quoted text
If the scope is working after you get the low voltage power supplies in spec, you can use it to measure the ripple. The test points are all called out in fig. 8-10 in the manual. Regards, Tom On 9/25/2020 7:09 PM, gary@... wrote:
Tom, --
This email has been checked for viruses by Avast antivirus software. |
|
Tom,
I do not have a HV probe. So are you suggesting, therefore that I leave the +55V adjustment alone? I think I read in the manual that changing this adjustment will require full re-calibration of the scope. Is that really the case? I am not using this scope for critical situations, but would like to perform up to it's potential. |
|
Don't worry too much about the HV, it's most likely going to be good once you set the +55 volts to the right point. There is no HV adjustment, only a check of the cathode bias level (-2450 volts). Remember, the last time it was calibrated, the 55 volts was the first thing set. You could do the crt grid setting though. See the adjustments section.
toggle quoted message
Show quoted text
What other test equipment do you have access to? A signal generator, a pulse generator?... Any nearby friends with some good calibrated equipment? Good luck, this was one of the Tektronix great scopes. On 9/25/2020 8:27 PM, gary@... wrote:
Tom, --
This email has been checked for viruses by Avast antivirus software. |
|
I adjusted the +55V to exactly +55.0, and the other voltages fell in line. I attempted to measure the ripple with the scope, but not sure of the results. At the +110 TP I think i was getting about 15mV, vs. a spec of 20mV. At the lower magnitude voltages the ripple was to small for me to determine a value, but if i had to guess it looked to be lower than spec. Not sure I was set up right, so I measured ripple on a 12v wall wart, as a test, and those measurements seemed to be working.
So let's assume my PS low voltages and ripple are OK, or are at least not so out of wack that I do not need to replace the PS caps - for now. I am still getting the compressed trace, intermittently. As far as other equipment, I do not have a function generator, only an audio frequency generator that to up to 100k Hz, and I am in the process of overhauling that. Before I call in favors, I would like to do some more checking. Anyone have any ideas? Thanks, Gary |
|
(Sorry if these have already been suggested -- I haven't read the whole thread): Have you tried cleaning the timebase switch contacts? Does jiggling the switch have any effect on what you see?
toggle quoted message
Show quoted text
Also, I have (rarely) encountered self-resetting tantalums. A tant generally fails as a hard, almost superconducting, short, as I'm sure you know. But I've come across a couple that would intermittently short and sort of self-heal before ultimately shorting forever. It's a low probability phenomenon, so I'd only consider it after eliminating the usual suspects. Good luck! --Tom -- Prof. Thomas H. Lee Allen Ctr., Rm. 205 350 Jane Stanford Way Stanford University Stanford, CA 94305-4070 On 9/26/2020 10:56, gary@... wrote:
I adjusted the +55V to exactly +55.0, and the other voltages fell in line. I attempted to measure the ripple with the scope, but not sure of the results. At the +110 TP I think i was getting about 15mV, vs. a spec of 20mV. At the lower magnitude voltages the ripple was to small for me to determine a value, but if i had to guess it looked to be lower than spec. Not sure I was set up right, so I measured ripple on a 12v wall wart, as a test, and those measurements seemed to be working. |
|
What you have sounds like a classic intermittent.
toggle quoted message
Show quoted text
Thinking of the possibilities, and since the power supply voltages are constant. 1) intermittent contact. That could be a power supply into the board. 2) intermittent contact. That could be a transistor/IC in a socket. 3) intermittent capacitor. Not so likely unless it's in a supply filter, and I'd think once failed, it wouldn't open back up again. 4) intermittent transistor: Generally, they fail and that's that unless somehow they're leaky or noisy. 5) intermittent resistor: Can happen, you'd have to monitor voltages. 6) intermittent Pot: Some of the pots are direct gain or centering controls. 7: Intermittent contact: switches.... x1 x10 gain typically. Generally, you pull the parts that can be pulled, transistors, that is, and put them back in. Ditto for the connectors. For the pots, I'd look particularly at the gain adjust pots and look to see what voltages (you could measure with a meter, but even better would be the scope's vertical channel. (which is behaving). If you don't have a horizontal problem (and I'd use your function generator to put in a 1 Khz sine to the horizontal channel. Pick a point where you want to monitor, hook the vertical to there (adjust vertical gain for 45 degree slope), and watch what happens. If the point you monitor is changing, then the angle of the line won't change, but the vertical and horizontal size will. If the point doesn't change, then the angle will change and the vertical deflection won't. Mark the position of the gain pots, and tweak them very slightly to see if that solves the problem. Monitoring is your best option, right after making sure there are not intermittent contacts. Harvey On 9/26/2020 1:56 PM, gary@... wrote:
I adjusted the +55V to exactly +55.0, and the other voltages fell in line. I attempted to measure the ripple with the scope, but not sure of the results. At the +110 TP I think i was getting about 15mV, vs. a spec of 20mV. At the lower magnitude voltages the ripple was to small for me to determine a value, but if i had to guess it looked to be lower than spec. Not sure I was set up right, so I measured ripple on a 12v wall wart, as a test, and those measurements seemed to be working. |
|
Moving or jiggling the time base switch (Time/Div and Delay Time switch) has no effect on the problem. I have not cleaned it as it is difficult to get to, even after lifting the DM44 module.
Regarding the tantalum caps, I have read that these can be problematic, but not ready to start replacing parts that may not be related to the problem. However, I do restoration of 30-50 year old audio gear, and replacement of electrolytic caps is pretty much the starting point, so I understand replacing caps that simply due to age, are likely to be bad, marginal, or will go bad soon. However due to the complexity of the 465B, I am reluctant to take this apart unless there is a goal (finding my problem - something specific to check and either identify or rule out) -- perhaps that is too much to expect. However, the reason I am asking for your time and tapping the experience and wisdom on this group, is to proceed based on that wisdom, rather than blindly. If that experience indicates that I should start by replacing certain caps (e.g. the electrolytics in the PS, and some/all tantalums), then I will gather up the gumption to do so - and will need to ask more questions. But taking on this task and then finding I still have the compression problem, while giving me a more "stable" scope, does not really fix the problem. I would rather find/fix the problem, and then with the knowledge that my scope does not have a major deficiency, address the (potential) issue of the caps. Is this thinking backwards? Thanks to all for your patience and advice. Gary |
|
All I can convey is what I've learned from decades of fixing scopes (I own almost 200 of them, and perhaps a dozen or so 465/465B and related models). By far the most common sources of trouble are the two that I've mentioned: Flaky contacts (which afflict pretty much all vintage electronics), and tantalum capacitors. Sure, other things go wrong, too, but those two dominate by far, so I spend a little time exploring those first before moving on down the list.
toggle quoted message
Show quoted text
Tek seems to have purchased a bad (and large) batch of tantalum caps in that era, so when I see a failure on the output side of the PS, I no longer spend a lot of time checking. It's much faster to replace the tants on the afflicted rail(s). There are not that many, thank goodness, and they are easy to access and replace. If I see a problem on the input side, it's easy enough to see if there are failed rectifiers or a bad primary cap. The latter are a bit of a PITA to replace (certainly compared to tants), but still readily doable. But you aren't having a problem with the primary side of the PS, and it isn't even clear that you have a PS problem of any kind. I share your disdain for shotgunning. However the problem here is that you appear to lack the necessary instrumentation to perform more detailed investigations, and you are also reluctant to disassemble things much. That severely limits what you can accomplish. It's hard to advise you how to proceed, given the overly constrained set of steps you are willing/able to take. So, I'll limit my comments to what I would do, and you are free to ignore my advice, whether for taste, or lack of tools. Put the scope in x-y mode (to bypass all the sweep generator circuits) and drive both x and y with the calibration source on the front panel (it's just a handy signal source). If you still see width compression there, the problem is downstream of the sweep generator. If you no longer see width compression, then the problem lies before the deflection amplifier. Wherever the problem lies, I would use Deoxit (don't flood; a little will do a lot) on all connectors to/from the horizontal circuitry, and the pushbutton switches (these have often been a source of maddening problems for me). If the first step revealed a problem with the sweep generator, note that I have also sprayed Deoxit on the contact fingers of the sweep speed knob, despite warnings that it may be a bad idea (I have never done so on the vertical attenuator fingers, though, where the danger of affecting the transient response is much higher). Like you, the prospect of disassembling the whole thing to do it the factory-recommended way (sliding strips of paper soaked in IPA between each contact pair) seems unattractive. I do it anyway, and have not had any problems. Of course, I have the luxury of many spare scopes, so you may have a different tolerance for gambling. Proceed at your own risk. I would reseat all ICs on the relevant boards, and carefully inspect all pins and socket contacts in doing so. Depending on the environment in which the scope was used/stored, weird corrosion can occur. Carefully clean any contacts that look dodgy. If none of the foregoing fixes things, I would monitor the power supply voltages and watch for dips (problems from spikes are rare). I long ago jimmied up a simple comparator-and-latch gizmo that does the watching for me. Set the lower voltage limit (e.g., 20% below nominal; you're not looking for tiny effects here), hook it up to the rail in question, and go watch Netflix. If the supply ever dips below the set threshold, an LED illuminates (and stays on), letting you know that that rail is flaky. Replace tantalum. Repeat. You can monitor several rails simultaneously; the monitoring circuit is simple to build. Or just replace the tants on the board(s) in question. It won't take much time. Good luck. The 465B is a nice scope, and well worth fixing up. --Cheers, Tom -- Prof. Thomas H. Lee Allen Ctr., Rm. 205 350 Jane Stanford Way Stanford University Stanford, CA 94305-4070 On 9/26/2020 13:00, gary@... wrote:
Moving or jiggling the time base switch (Time/Div and Delay Time switch) has no effect on the problem. I have not cleaned it as it is difficult to get to, even after lifting the DM44 module. |
|
In case it helps...several years ago, after dealing with a failed tantalum cap...I compiled info for the 465B to make the next one easy to find.
It is in the files section of this forum but here is a direct link. It is sortable by clicking column headers. Good luck. /g/TekScopes/files/465Btants.html ignore the email address in it. I lost editing rights in the yahoo->groupsio move. |
|
Thanks very much for creating that doc -- I'm sure it has helped and will help many!
toggle quoted message
Show quoted text
-- Cheers, Tom -- Prof. Thomas H. Lee Allen Ctr., Rm. 205 350 Jane Stanford Way Stanford University Stanford, CA 94305-4070 On 9/27/2020 11:01, pdxareaid wrote:
In case it helps...several years ago, after dealing with a failed tantalum cap...I compiled info for the 465B to make the next one easy to find. |
|
Tom L.
Thank you for your very helpful response. Your advice is exactly why I came to this forum. I am proceeding slowly due to my relative inexperience with electronics. I am self-learning this craft (just as a hobby as I near retirement) by reading as much as I can, mostly older fundamental electronics materials that include vacuum tubes, since that is my main interest. I pretty much started at the beginning only a year or two ago. That aside, I am methodical, and check and re-check my procedure as all of this is new to me. The whole reason I have a 465B is my desire to learn, fix, and learn some more. I could have bought a new Rigol or Siglent, but don¡¯t regret the choice I made, as I understand and appreciate the quality of the 465B. That said, my electronics experience is analogous to having fixed a lawnmower, and by comparison the 465B is 12 cylinder Ferrari (out of my comfort zone to be sure, but not fundamentally different). My lack of other test equipment is temporary, I have a B&K 4040A 20MHz Sweep/Funciton Generator on the way, and I am willing to acquire additional equipment as needed for the project at hand. Regarding disassembling the unit, I have read the Maintenance Section 5 of the manual and think I can handle it, but when I do, I want to have needed parts on hand to replace the usual suspects, the tants and electrolytics and do more cleaning. One note on cleaning, I do not see any socketed transistors or chips ¨C others have made reference to cleaning the sockets. Are sockets used in the 465 and low SN B¡¯s, but not >60k 465B? Everything looks soldered to the board to me. I will post what I have tried by way of diagnosis separately to keep this from getting too long. |
|
Over the past week I spent some quality time inside the scope. Here is what I did and learned:
? Compression of the trace occurs in XY mode (Ch 1 and Ch2 probes connected to the calibrator) and with input from the external A Trigger input with a 50 ohm BNC cable to a clip on the calibration loop. Given this, based on your advice, and understanding it based on the block diagram, I focused on the Horizontal Amplifier circuit. ? I examined the schematic of the Horizontal Amplifier (Diagram 10), and the related components, which are on board A4, which is the large board on the bottom of the unit, and therefore very accessible. There are no tantalum caps in the horizontal Amp circuit. There is one electrolytic (C4461, which I will replace when I get one). ? I used my DMM to take DC voltage readings at test points 85 through 90. The largest variation I read from the spec was 10% at TP 85, the others were between 1.6% and 7.8% of spec. ? Since I do not have a second scope, but do have 2 DMMs (a poor man¡¯s 2 channel scope?) I measured AC (RMS) at these test points and observed what happened to the AC readings at each TP pair when the trace collapsed. o At points 85/86 which are near the beginning of the Horiz Amp circuit, there was no change to the AC voltage when the trace collapsed, but at the next two pairs (TP 87/88 and 89/90) the AC dropped significantly (by 75% to 80%) when the trace collapsed. o This seems relevant to me, but I don¡¯t know if this is a concurrent symptom of the trace collapsing, or if it means that perhaps the problem is between TP 85/86 and TP 87/88. o This portion of the circuit includes several potential culprits: x10/x1 Mag switch and related pots (mentioned as a potential issue by Shailendra), and two pairs of transistors. o I then tried something simple that I should have tried at the beginning ¨C I pressed in the x10 Mag button. The trace expanded to both edges of the screen and beyond, as expected. But I am pretty sure, there is NO compression in x10 mode. I even counted the number of square waves in the compressed x1 trace vs. the x10 and the same number of waves are present when scrolling horizontally. o I then cleaned the x1 and x10 Gain pots, Mag Registration, and X gain pots with no visible effect. ? One other item to note, tonight for the first time the trace is now always in the compressed mode, not intermittent ¨C I can get it to slightly expand by manipulating the horiz position control but it jumps back to compressed. The only way I can get it to expand and hold is with the x10 button pressed in. That's as far as I have come. ? Any other suggestions for trying to locate the problem within the Horizontal Amp, are there other areas I should check? Or am I at the point that I order caps, and replace the tants and lytics. Since there is not much in the horiz amp circuit, should I go ahead and order the transistors and be ready to replace any that test bad? Should I preemptively replace transistors? ? My very next step is to put together a list of caps (Thanks very much to pdxareaid for the list of low SN tants. I checked it and it appears to match the listing for the high SN units too). But now I need to translate this list of what is in the unit into specific caps to order ¨C selecting the right voltage (go up a bit?), lead configuration, and tolerance (rated at 20%, should I go to 10%?). I find this process very tedious, but necessary ¨C does any one have a BOM with MFG or Mouser part numbers? If not, that will be my contribution to this forum ¨C after I ask you guys to check my work. I would also like to have the pots on hand, and looking them up tonight. Sorry for the long post, just trying to help you help me. Thanks for being patient with a rookie. |
|
The Tek scopes are, at that level, more like 8 lawnmowers bolted together than one gadget that is 8 times more complicated than a lawnmower.? The scopes are rather modular, and there's a very good description of how the individual sections work.? so: using the manual, isolate a problem to the section, then go through the troubleshooting information.
toggle quoted message
Show quoted text
GIGANTIC HINT:? Make sure that the power supplies are accurate before trying to fix anything. Harvey On 9/30/2020 11:33 PM, gary@... wrote:
Tom L. |
to navigate to use esc to dismiss