|
Happy New Year everyone!
I have a 7904 s/n 215841. I got it at an auction in the UK where I live, where there don't seem to be many of this series available, and it was FILTHY when I picked it up. Knowing the tantalum issue I immediately replaced as many of the PSU tants as I could get at without dismantling the unit completely, and removed as much of the grime as I could which was like oily black soot and covered all the circuit boards - got a surprise when I discovered that the HV line was bright pink! Now it works, the PSU comes on and the voltages are within spec, screen is bright (albeit not a patch on my old 545 - is this normal?).
The issue is that almost immediately there is a slight ticking noise inside the PSU. It starts as a fairly regular tick (every second or two) but after ten or twenty minutes has become more of a quiet crackling noise. This is far lower in volume than eg. a start up tick. There was no visible light from arcing when looking in a darkened room but I didn't want to stick my face too close, also there is no discernable burning smell. After the crackling develops the voltages remain steady, the screen doesn't change. I dug into the PSU, lord what a job, to look for anything I could find but apart from plenty of grime didn't find any tell-tale carbon arcs or blackened components. I had a look inside the cap multiplier but, again, no obvious arcs. The neon that is on the inverter board had stopped working so I replaced that, changed out the 470 Ohm sense resistors R1429 and 1439 which were burnt but still near-ish to spec, reassembled, and tried again, more in hope than expectation. No change. My questions are:
1. Can anyone point me at the likely source of the crackling? I am fearing it might be the transformer futzing out, but have no idea, after pulling the PSU apart, how I could test it reliably (and safely). If you think it is that, can it be rewound? I can't do this sadly, but I do know a couple of professional trafo winders. If the symptoms scream "cap multiplier" at you I do have a used spare that is labelled good, bought from Sphere's old stocks but, before replacing it haphazardly, I wanted to ask those with more experience than me how to proceed. That's all of you I suspect...
2. The service manual suggests testing impedances on the LV regulator board test points with the power wiring looms connected and disconnected , however their advice is to investigate if the measurements are "significantly lower". But these are for s/n 260000 and up. My PSU is not much different to the circuits in the Tabalabs .pdf. However, my measurements differ from the specs given in table 4-3 as follows:
+50V TP with wiring loom connected - spec 3.4k Ohm, measured 2.95 kOhm. Is this significantly lower? Disconnected, the regulator resistance rises to 23.6 kOhm.
-50V TP with wiring loom connected - spec 270 Ohm, measured 3.74kOhm. Disconnected, the resistance is 4.34 kOhm. This appears significantly higher but is it an issue. Thoughts?
3. The manual suggests testing the Pre-reg power supply across R1287. The waveform is more or less like that illustrated in 4-4 (A) showing correct operation. Looking at the burst voltages at the test points given in table 4-4 there are stable voltages with little evidence of bursts, and they come at 25 kHz rather than 100 Hz intervals, I assume this is normal?
4. The new neon bulb on the inverter board is supposed to blink to indicate charge on the reservoir caps. Mine flickers dimly and continuously. Is this normal or does it indicate something wrong?
Any and all help, advice, and pointers gratefully received.
|
I have a 7904A with similar issues on the bench also with a VERY early serial number 507. In my case I have confirmed that there is issues with the main bulk filters they are arcing internally. 200 Vdc cap is being fed by a 100ma 200 Vdc current source is building to about 70 ish Vdc then arcing down to about 50vdc. It Is causing a static like tic when it arcs over. This one is blowing the mains fuse.
For the resistance testing these measurements are not an exact spec but more like a gross find the short. This is due to the meter used for the measurements. All Ohms ranges have a voltage and current applied to the circuit and this can throw off the readings. Some meters top out at 3 Vdc some like my Keithly go up to 10. There is no way of knowing what active components are turning on at the higher voltage. Exact is not needed there but in the ball park is fine.
25Khz is the switching frequency of the supply.
4 is the power switch good? The neon should stop eventually it can take a bit given the health of the bulk caps if the power switch is leaking current, it will keep the caps charged due to line -> switch -> rectifier -> bulk caps. This would however be a REALLY odd failure mode.
Zen
toggle quoted message
Show quoted text
-----Original Message----- From: [email protected] < [email protected]> On Behalf Of James via groups.io Sent: Sunday, January 5, 2025 9:34 AM To: [email protected]Subject: [TekScopes] 7904 pre-260000 PSU crackling Happy New Year everyone! I have a 7904 s/n 215841. I got it at an auction in the UK where I live, where there don't seem to be many of this series available, and it was FILTHY when I picked it up. Knowing the tantalum issue I immediately replaced as many of the PSU tants as I could get at without dismantling the unit completely, and removed as much of the grime as I could which was like oily black soot and covered all the circuit boards - got a surprise when I discovered that the HV line was bright pink! Now it works, the PSU comes on and the voltages are within spec, screen is bright (albeit not a patch on my old 545 - is this normal?). The issue is that almost immediately there is a slight ticking noise inside the PSU. It starts as a fairly regular tick (every second or two) but after ten or twenty minutes has become more of a quiet crackling noise. This is far lower in volume than eg. a start up tick. There was no visible light from arcing when looking in a darkened room but I didn't want to stick my face too close, also there is no discernable burning smell. After the crackling develops the voltages remain steady, the screen doesn't change. I dug into the PSU, lord what a job, to look for anything I could find but apart from plenty of grime didn't find any tell-tale carbon arcs or blackened components. I had a look inside the cap multiplier but, again, no obvious arcs. The neon that is on the inverter board had stopped working so I replaced that, changed out the 470 Ohm sense resistors R1429 and 1439 which were burnt but still near-ish to spec, reassembled, and tried again, more in hope than expectation. No change. My questions are: 1. Can anyone point me at the likely source of the crackling? I am fearing it might be the transformer futzing out, but have no idea, after pulling the PSU apart, how I could test it reliably (and safely). If you think it is that, can it be rewound? I can't do this sadly, but I do know a couple of professional trafo winders. If the symptoms scream "cap multiplier" at you I do have a used spare that is labelled good, bought from Sphere's old stocks but, before replacing it haphazardly, I wanted to ask those with more experience than me how to proceed. That's all of you I suspect... 2. The service manual suggests testing impedances on the LV regulator board test points with the power wiring looms connected and disconnected , however their advice is to investigate if the measurements are "significantly lower". But these are for s/n 260000 and up. My PSU is not much different to the circuits in the Tabalabs .pdf. However, my measurements differ from the specs given in table 4-3 as follows: +50V TP with wiring loom connected - spec 3.4k Ohm, measured 2.95 kOhm. Is this significantly lower? Disconnected, the regulator resistance rises to 23.6 kOhm. -50V TP with wiring loom connected - spec 270 Ohm, measured 3.74kOhm. Disconnected, the resistance is 4.34 kOhm. This appears significantly higher but is it an issue. Thoughts? 3. The manual suggests testing the Pre-reg power supply across R1287. The waveform is more or less like that illustrated in 4-4 (A) showing correct operation. Looking at the burst voltages at the test points given in table 4-4 there are stable voltages with little evidence of bursts, and they come at 25 kHz rather than 100 Hz intervals, I assume this is normal? 4. The new neon bulb on the inverter board is supposed to blink to indicate charge on the reservoir caps. Mine flickers dimly and continuously. Is this normal or does it indicate something wrong? Any and all help, advice, and pointers gratefully received.
|
Thanks Zen, much appreciated.
I checked the main caps fairly cursorily for capacitance, which was textbook value, but didn't check them at proper operating voltage for issues. I will certainly do so when I pull the PSU apart. I forgot to mention that the crackling continues (decaying) for a minute or two after switch off, maybe that helps narrow the culprit down?
Thanks for the info about the resistances, understood!
Re the neon, my poor description, it flickers dimly (like most neons do) when the psu is switched on and very quickly stops after it is switched off. It doesn't blink though if that matters. I note that the reservoir caps don't hold charge for any time after switch off. The switch, like all Tek switches, looks built to last forever. It functions fine thankfully.
My understanding is that the PSU in the 7904A is rather different but have no experience to back that up.
|
On Jan 5, 2025, at 06:34 , James via groups.io <james@...> wrote:
The issue is that almost immediately there is a slight ticking noise inside the PSU. It starts as a fairly regular tick (every second or two) but after ten or twenty minutes has become more of a quiet crackling noise. Are you sure it's in the PSU? Does it change if you bump the HV lead? I've had a tic from my 7704A that sounds like a static crackle if I move the HV lead with a nonconductive tool. Have not been able to make that go away. I have a 7904 that's a bit older than yours, and IIRC it had a crackle that went away after disconnecting the CRT anode connector, shorting it to the frame, cleaning it, and applying some silicone grease. It still has a PSU tick when starting, but I understand this is common. Adam
|
Take a long plastic tube, at least 1 foot.? You could use cardboard, but plumbing material is fine.? Stuff a microphone down one end and seal it in with tape.? Run the microphone to an amplifier.? You now have a sensitive directional sound probe. Putting the end on the suspected part should tell you a lot.
You can also use a stethiscope? with a long piece of tubing (plastic!) to do the same thing.
Harvey
toggle quoted message
Show quoted text
On 1/5/2025 10:56 AM, James wrote: Thanks Zen, much appreciated.
I checked the main caps fairly cursorily for capacitance, which was textbook value, but didn't check them at proper operating voltage for issues. I will certainly do so when I pull the PSU apart. I forgot to mention that the crackling continues (decaying) for a minute or two after switch off, maybe that helps narrow the culprit down?
Thanks for the info about the resistances, understood!
Re the neon, my poor description, it flickers dimly (like most neons do) when the psu is switched on and very quickly stops after it is switched off. It doesn't blink though if that matters. I note that the reservoir caps don't hold charge for any time after switch off. The switch, like all Tek switches, looks built to last forever. It functions fine thankfully.
My understanding is that the PSU in the 7904A is rather different but have no experience to back that up.
|
Clew...scope, bought in filthy condition....
Crackling...
Almost certainly the HV ultor 21 KV at X7 multi
On our 7904A, exactly same problem. Doubtful if LV PS or bus caps.
We have experience in HV since 1970s. Suggestion, open left scope cover, in a completely dark room, look for corona at the rear upper left, th HV doghouse hole around the large HV connector. Smell for ozone.
We opened up the doghouse, unplugged connectors, applied isopropyl alcohol on cotton with longnose pliers to entire HV cable, connector and CRT cap.
dry with gentle heat, carefully reassemble. May have to sit overnight for complete drying.
Crackling was gone. Hope this helps you
Jon
|
Hi all and thanks for the replies and help. I'm pretty sure it's in the PSU but I will reconnect it all (again), fire it up, and try the trick with the microphone in a plastic tube. Trouble is, I'm 90% certain it comes from inside the PSU and when I had the scope turned on in a darkened room I couldn't see any corona. My sense of smell isn't the best but I detected no ozone. But I will try again and report back.
I did quite thoroughly clean the HV lead and the connector originally, and cleaned the cap multiplier lead when I had the PSU apart. They were totally blackened with this oily grime, as were all the wiring looms. I imagined that this could lead to arcing and other unwanted stuff, so it seemed a good way to start. It's quite astonishing that the thing worked at all!
What is the dog house hole? Is it the slot in the middle of the PSU panel where the HV lead and other wiring comes out? Once I have that confirmed I might try getting another plastic tube and seeing if I can sniff ozone through that?
Re bumping the HV lead, yes, I have tried it. No change, sad to report.
Currently waiting on a supply of tantalums and other bits to replace those present on the other boards, also a tool for inserting the connectors into new molex connector plugs before the connectors totally fall apart (the blue ones are already partially absent, I'm concerned for shorts).
|
The neon functioning is acting normal for operation of the PSU. The service manual calls it out as a "safe handling" indicator IE don’t handle the supply while it is still flashing. The supply has bleeder resisters for the larger caps so it sounds like they are functioning. That is what is helping with the quick drain off.
In my case I tested the caps out side of the supply with an SMU so I am 100% sure the fault is internal to the 980uF 200VDC though yours are not indicating like mine did one was reading higher then expected capacitance 1086uF with a climbing ESR 0.06 retest to 0.56
How do the spark gaps look? There are some glass tubes that looks like small fuses but are designed to arc over if they need to. If any are carboned up or dirty that will lower their arc over voltage.
If we are sure it is not the HV section then I would look for an arcing component. I have seen tants die that way when they cracked open some times you can even see the static. Or some for of contaminants.
If you have a fast enough meter are any of the power rails jumping around in time with the crackle?
If it is true arcing then an AM radio will be sensitive as a detector as well. Louder = closer
Absolute worst-case would-be breakdown in the transformer.
Zen
toggle quoted message
Show quoted text
-----Original Message----- From: [email protected] < [email protected]> On Behalf Of James via groups.io Sent: Sunday, January 5, 2025 1:13 PM To: [email protected]Subject: Re: [TekScopes] 7904 pre-260000 PSU crackling Hi all and thanks for the replies and help. I'm pretty sure it's in the PSU but I will reconnect it all (again), fire it up, and try the trick with the microphone in a plastic tube. Trouble is, I'm 90% certain it comes from inside the PSU and when I had the scope turned on in a darkened room I couldn't see any corona. My sense of smell isn't the best but I detected no ozone. But I will try again and report back. I did quite thoroughly clean the HV lead and the connector originally, and cleaned the cap multiplier lead when I had the PSU apart. They were totally blackened with this oily grime, as were all the wiring looms. I imagined that this could lead to arcing and other unwanted stuff, so it seemed a good way to start. It's quite astonishing that the thing worked at all! What is the dog house hole? Is it the slot in the middle of the PSU panel where the HV lead and other wiring comes out? Once I have that confirmed I might try getting another plastic tube and seeing if I can sniff ozone through that? Re bumping the HV lead, yes, I have tried it. No change, sad to report. Currently waiting on a supply of tantalums and other bits to replace those present on the other boards, also a tool for inserting the connectors into new molex connector plugs before the connectors totally fall apart (the blue ones are already partially absent, I'm concerned for shorts).
|
James,
The neon, DB1219, should stay lit for a bit of time after turning the unit off. By going off fast, that sure sounds like your main filters are bad. In these in my 7000 series, I used Nichicon LGR 1200mfd 250V to restuff the cans. I also replaced the 150,000 ohm 1W with 100,000 ohm 5W, mounted off the board a bit to allow for more air flow around it, to bleed the B+ faster. With the new guts, the neon takes 30 seconds to stop flashing after turning it off. Check the B+ across the cans or across the 150,000 ohm 1W resistor. It should be 310V nominal. If about 280V or lower, the cans are going/are bad. In mine, I put in a CL60 or 70 off the mains to reduce the inrush current further. Being 50 years old, it would be best to recap the piece. Higher value electrolytics will work - especially on the 50V supplies (raise from 8,2/10mfd to several times the amount of capacitance and use 100V like Nichicon UHE). The originals are likely 60-70V. For the 17V supplies, use 35V. For the 15V lines, 25V. For the low value caps, 1, 2,2mfd for example, I used film types. I did raise the wattage of some resistors in the scopes.
I wonder if you still have the oily grime in areas that are a resistance path causing the crackling sound. The boards and other areas can be cleaned with ammonia or some detergent like Dawn, rinsed with warm water then left to dry for a while. A fan blowing on the pieces and/or over a heat register from a floor vent overnight. The dry heat that long will dry enough.
There could be a gap where the ultor lead goes in the glass at the front of the crt. If so, RTV works to seal that.
Mark
|
One of my 7904s just crapped out with HV arcing inside the PS unit - very loud and obvious. I had just fixed it (it had other problems not HV) back in November, and I'm convinced the failure was due to high humidity damp and cold conditions out in the garage.
If your new unit is quite filthy from what you've seen so far, the inside is probably pretty bad too. If it's working now but making discharge noises, you should open it up and clean the HV section (and the rest too as needed) thoroughly, then look for evidence of arcing or component damage. Chances are it should be good to go after cleaning.
I think that mine unfortunately ended up damaging something. I should have shut it down immediately but I thought it might clear by itself. No such luck. It still showed a messed up display at first, but then went blank - bad idea to let it go.
Ed
|
James: "What is the dog house hole? Is it the slot in the middle of the PSU panel where the HV lead
The HV section has a metal cover (doghouse) with a 3/8" hole where HV U;tpr 21 kV connector exists from HV multiplier.
Adding insulation is useless.
Just be sure the cable and Alden connectrs are clean and ry.
Ozone: even0. 1 PPM is easily penetreating smell
No tube needed.
Jon
|
Jon, Ed, Mark and everyone else, thank you so much! I had to go to work today, an "occupational hazard".
Tonight I clarified a couple of things. The neon stays on for a good 30 seconds, but I need a dark place to see it in, so my bad that I didn't see it before, apologies. Second, the noise comes from the middle of the PSU. I'd love to say from the trafo or the caps but it's too tight in there to tell, definitely in there. Sounds like both ends of the unit. Sorry for the lack of certainty. Voltages say tight, the ripple is of variable frequency across voltages if that matters, this with a vertical and horizontal unit inserted. There are snaps, crackles, and pops from the hv cable when wiggled so this needs further cleaning as well, but it sounds different.
|
Oh, no smell of ozone either Jon. The trafo runs at 51 degrees, the caps at 31, this relative to a room temp of 20C, all limited by DIY equipment...
|
Hi all, I had some time so dug into the PSU a bit, swapped ot the last tantalum in there, and upgraded the aluminium 'lytics as per revisions list in the manual. The big storage caps I ran up to 195V for half an hour and checked for leakage, it was 10uA on one and 20uA on the other, they measure at 980uF and 1025uF, they seem absolutely fine to me. The HV box showed carbon or soot near the dog house hole and a mark on the insulating washer that wraps around that hole. There was also quite a bit of carbon/soot around the small hole in the aluminium case that could have been conductive? There is also a nick on the HV cable which doesn't go through the insulation, I could see no light near that nick when the PSU was working and crackling. So saying, I have no experience of coronas and arcs, and little of these kinds of voltages, so I have added some photos at the following link titled 7904 PSU crackling noise: /g/TekScopes/album?id=300059I have cleaned the box thoroughly and there was no evidence of any damage to anything caused by any arcing, nor is the mark on the washer still visible, but I don't know if there would be visible damage? There were also some light carbon deposits on the large ceramic caps. Is this a cause for concern? The little piece of loose insulation visible is (I think) probably a result of my bending the caps out of the way to get at the transformer wires to unsolder them. Please advise what you think? Might I have found the smoking gun by any lucky chance? Finally, the transformer which I expressed concern about is at last exposed. It doesn't look too bad to me, not blackened or anything, and no more filthy than most of the rest of the unit. The core is quite loose on the windings but there isn't anything obviously amiss to my untrained eye. Whilst the thing is apart, are there any useful tests I can carry out on the trafo to ascertain if the crackling might have been coming from it or is it rather unlikely that this might be the case? Many thanks for any and all pointers as always!
|
Hello again
Last two photos show Carbon and dirt accumulation on HV box , cables and Kapton film insulation
Needs through cleaning with 99% isopropyl alcohol.
Where is the HV cable nick, photo seems fine.
DO not fool with the HV transformer.
Very easy to damaged and costly to replace
Reassemble carefully, and test after cleaning
Jon
|
Thanks for the reply Jon,
The nick on the HV cable is about half way along from the dog hole to the connector, visible on the third image.
If I don't need to mess with the transformer I won't, it's just that I'm really keen not to have to reassemble and find the thing is still crackling because I didn't do all that I could when the PSU was apart!
|
Hello, I had a similar crackling and hissing problem with my 7904 years ago. It turned out to be the voltage multiplier U1615, internal arcing. The only visible indication of something being wrong was the burned resistor R1604, 13ohm. After replacement the scope was working again.
G?ran
|
I guess something to realize is that an air gap from the HV lead core to the outside of the insulation isn't enough to prevent arcing or even flash over; The dielectric strength of air is far less than that of the lead's rubber insulation.
As an example, I once repaired an arcing cable by filling the area where the (red) lead enters the socket in the middle that runs past the chassis by making a gradual "glove" out of silicone rubber from the cable to the white socket. And yes, everything was clean but arcing slightly before that!
Raymond
|
Raymond the 21-24 KV CRT Ultor is polyethylene as is,the molded Alden connector.
Indeed any nicks or splitting of the cable will cause arcs or corona to the nearby grounds og the ch?ssis , doghouse cover, etc.
I would try to fix the fault on the cable with heavy duty shrink tube or high grade electric tape.
Jon
|
Jon,
Just to prevent a possible misunderstanding: I wasn’t talking about the ultor but the white two-part connector somewhere near the middle of the HV lead, where the (usually) red lead enters it.
Personally, I strongly dislike using tape to insulate a cable/wire. With the HV lead, I don’t think shrink tube exists that goes around either side of the lead and shrinks snugly around the lead, without disassembly, that is.
The silicone rubber becomes an integral part of the lead and looks nice because it tapers between lead and connector.
Raymond
|
Jean-Paul
G?ran Krusell
Raymond Domp Frank