How to Work on the LV Supply on a 533A


 

I have come to the point where I need to examine the capacitors in the LV power supply of my 533A. The CRT has a scratch on the inner phosphor surface, which makes scared that there might be something broken loose inside, so I’d like to remove the CRT before manhandling the rest of scope to get at the bottom circuitry. When I removed the CRT I stored it on its face, but I don’t want to do that if there’s a loose part inside that would cause further damage to the face. Is it safe to rest the CRT on its side if I provide bracing in the form of cardboard or closed cell foam supports?

I am going to be investigating the components around R760 in the 500 V supply. R760 was vaporized so I’m expecting that the capacitors in C760 may have failed and are now behaving as conductors, but I’m open to other suggestions for things to check. The manual says that I should be checking the resistance from the power rails to ground, which I assume should be effectively infinite, if everything were working. Is this correct?

— Jeff Dutky


Wayne
 

I would not expect infinite resistance to ground from the rails. There are sometimes voltage dividers from power rails to ground to establish levels for various functions in circuits.


 

So the manual only says “ check the resistance values between the outputs of the Low-Voltage Power supplies and ground,” but does not seem to have a list of what those resistances should be (it’s not a long manual, it it’s always possible that I have missed something). Unless someone can point me to the documentation telling me what resistances to look for I will simply measure the resistances and post the list here.

— Jeff Dutky


Mark Vincent
 

Jeff,

Setting the crt on its side is fine. I have done this many times. Put it in a place it will not be disturbed while it is out.

With R760 open, the can is shorted. Restuff it with new ones that will physically fit inside the can. Unlikely will check shorted on a meter. Having seen this enough times in the past, the short will be at some voltage higher than a typical meter and the B+ it sees. Being the same age, the other cans should be restuffed. I have a 535A I restored. The cans in mine are now restuffed with new high temp, low ESR long life types. I have restuffed so many cans, I do not mind. I find it part of the job. The two 150mfd can in parallel can be a single 330mfd (LGR for example) or larger in one can with the other disconnected from any voltage and lugs shorted. The unused can with the lugs shorted will be a visible note that it is not electrically in circuit, only physically there. The ,01mfd films should be replaced with new 600/630V types. The 533/A has a ,022mfd for one and that should be replaced.

Check the diodes for the 500V supply. One or more may be shorted now. If one or more is/are bad, it is best to replace the four. You can add NTCs to the primary and secondaires to reduce inrush current. The NTC resistance I use on the primary is 16 or 10 ohms and something 20-30 ohm NTCs for each secondary. The voltage drop is too small to be concerned with.

You may look at the primary voltage taps being used. Line voltage has increased since they were made. You may set the primary to the higher voltage primary taps.

Mark


 

Mark,

Thanks for the complete rundown. I have questions about how to restuff the cans: how do you get the cans open? Do you cut the cans at the base and reattach them later (with solder? something else?), or do you uncrimp them?

I tried to check the diodes in-place and got confusing readings. Do I need to lift them to get good readings, or can I tell if they're shorted with them still in circuit? If they are shorted, what do I replace them with?

I know, lots of questions. I've only been at this hobby for less than a year, and I feel like I still know next to nothing.

Thanks.

-- Jeff Dutky


Harvey White
 

If you're going to eviscerate a canned capacitor, you might want to consider a pipe cutter once you remove it from the PC board.

Unless, of course, you have a lathe and a cutoff tool.

Harvey

On 6/5/2021 8:16 PM, Jeff Dutky wrote:
Mark,

Thanks for the complete rundown. I have questions about how to restuff the cans: how do you get the cans open? Do you cut the cans at the base and reattach them later (with solder? something else?), or do you uncrimp them?

I tried to check the diodes in-place and got confusing readings. Do I need to lift them to get good readings, or can I tell if they're shorted with them still in circuit? If they are shorted, what do I replace them with?

I know, lots of questions. I've only been at this hobby for less than a year, and I feel like I still know next to nothing.

Thanks.

-- Jeff Dutky





Dan G
 

Hi Jeff,

Based on the pictures you uploaded recently, this 533A unit seems to be
wonderfully preserved in its original state.

Before embarking on an adventure of massive indiscriminate destruction of
its original components, perhaps you may consider testing the capacitors first.
Since they have not been used for decades, the aluminum oxide dielectric
has probably un-formed, and the breakdown voltage will be much lower than
spec. Only one leg needs to be unsoldered, and then you can test for DC
leakage. (A curve tracer is very handy for this.) If the leakage is high, it is
generally possible to re-form these capacitors. I have a number of Tek
instruments from the 50s and 60s, all with original can capacitors, and they
all work perfectly fine.

Something tells me that a few decades from now, these old cans will still be
re-formable and usable, even after the modern replacements have failed
and leaked their corrosive contents all over the instrument.


dan


Mark Vincent
 

Jeff,

You are the first person to ask how I restuff cans in the years I have done this. Others wanted to know yet did not want to stay to see how it is done.

I uncrimp them by a pair of dikes a bit at a time until the end is vertical. Make sure to get the neg. tabs flat before removing from the plate to ease removal. Then I heat them with the top of the can(s) is/are on the eye of a stove on low or a bit above for 15-40 minutes. There is no way to tell how long it takes for them to be hot enough to pull the guts out. I use an oven pad to hold the can then I use vice grips on the lugs after I think it is hot enough. If it is ready, the guts will come out. The tar inside is scraped out by a small flat blade screwdriver. Do not worry about any leftover tar. Get as much as practical. The guts are cut at the inside of the wafer and discarded. The negative ring is pulled from the Al tab. The front plate is removed from the black rubber like substance. I use a toothbrush and Dawn to clean the white part and other liquid out then let dry some by blowing a fan on them or something warm while I do what I need to with the new condensers and maybe scratch the ring enough to get solder to stick on the flat outside. They will be dry enough by time you finish this. The tabs are removed by needle nose pliers. The tabs are broken off at the crimp. The inside part of the slots the tabs came out of are carved into part of an arc to allow the positive lead of the new lead to come trough. An awl or ice pick is pushed through the inside plate and black part to enlarge the holes for the leads. Depending upon the number of condensers are used inside, the negative lead(s) will come out in a notch cut bit by bit to keep from breaking the phenolic base that is on the outside. A file can be used. The notch is also done thought the black and inside phenolic plate. If the new condensers are able to be stacked in the can, add solid wire leads on the leads to make them long enough to come through the bottom while making sure the stacked parts fit inside the can.. For a snapin type this method has to be done to get the leads out to solder to. An example is the 330mfd or larger to replace the two 150mfd cans. Put the positive lugs back in by hand through the D shaped holes/slot. The positive leads are inserted through the "sandwich" long enough to go through the solder hole and bent back to hold and larger solder contact. The negatives will be in the notch(es) which will be between the ring and inside of the can. You will see the dent by the lead(s) when finished, do not worry. Put the negative ring on with the tab back in the notch. Make sure to orient it so the D, [], > and _ are going through the plate to the wire(s) they are supposed to. The new guts are still exposed. Once orientation is done, then the can is put over the new guts. I make sure the numbers on the can are oriented so that they are seen easily, or as best possible, when they are installed and the part that is removed (side panel, bottom plate, back, etc.) when servicing is off. I do this to know what the value and voltage of the insides are without using a small mirror and fighting the other guts to see the characters on the can/paper tube. The negative lead(s) will be bent over the ring and between the bottom part of the can, trimmed then soldered. Once the soldering is done then the part that was bent out is now bent back. I will use dikes to bend it at a few points to start the bending and holding the guts in. I then tap the part where it will bend back on the ring at 45deg. angle with a hammer on the floor to have surface that is solid and will not be damaged like a wood workbench. The can is horizontal doing this. Once it is bent all around then I put it on its top and tap a little harder making it close to the original arbor press in appearance and making sure the negative lugs are able to be twisted, maybe, and soldered to. This part is pressed against the plate so make sure to get it as flat as practical. For a metal plate I leave the neg. lugs alone and solder them to the plate to prevent bending again. For phenolic, bend enough to hold the can. The plastic support and tube used on some at the base and side will fit the way I restuff. I hope I did not leave out a step. It sounds daunting. It is more messy and laborious than anything. The first time you do this, it will take the longest time. As you do more, you will get faster due to being confident and remembering what to do. For the small cans, 22mm is the max I.D. available, e.g. a 10 and 12mm dia condenser will fit side by side. For the larger ones, 32mm is the max I.D.. Sizing the new ones can be the longest in time to source. Using larger capacitance than original is fine. You can have a the new guts in a triangle at times.

The cans on some are the turret type molded in a hard plate. Holes will need to be drilled near the turrets for the leads. Examples of this type are in the 453/4/A types.

The screw lug cans are another way to restuff. The 7000 series I have done these with the original 550 and 950mfd replaced with 560 and 1200mfd respectively.

I figured out how to restuff before this group existed. I looked at what I wanted done then figured out how to do it. No one told me how. I like to keep things original looking. I have not and will not restuff an old wax condenser. That is too much trouble and I do not like the wax outside continue to drip when it is warm. That would be wanting original being carried too far. Resistors are another thing. I have already said what I use and why. I will stick by my belief in doing a job right to begin with. The resources to go back outweigh doing a job "just enough" the first time to save resources. More to begin with can save even more later. Time and aggrivation have value.

I would suggest raising the wattage of some resistors. If you want the R numbers, ask. I will use the 535A list I have when I did mine and compare it against your model to see if there are any differences in parts.

Mark


 

Mark,

Thank you for the detailed explanation. I will try this on a couple of dead caps that I have from other Tek instruments before I tackle the 533A, just to get the lay of the land. For most of my Tek equipment I am unconcerned about making the instrument look “original” but this is going to be a shelf queen (unless it turns out that I find a good use for it, which is not beyond the realm of the possible: it came with a Type D plug-in, which makes it my only scope that can do differential measurements), so cosmetic details matter. For the scopes I intend as daily drivers, or the ones that have sentimental value , I’m much more concerned that they be working, rather than that they look original. The other scopes also happen to be of later vintage than the 533A (from the 70s and 80s) and it’s a lot more work to get a look at the innards.

What is your policy on resistors? I wound up replacing the 10 ohm R760 with a vintage carbon comp resistor of the same wattage because a) I could hand select one of the right resistance value, and b) I interpret the low value as being essentially a fusible element, meant to protect other components in case of abnormal operation or failure. Also, I like the look and feel of the old A-B carbon comps, even if they tend to drift: nice sharp edges on the cylinder, nice heavy wire for the leads.

For the diodes I am looking at spending real money on some top-hats that are exact matches for what’s in the unit now (1N3194s). Again, these are easily visible under casual inspection, so “original parts” have some value, but I’m open to substantive arguments to the contrary.

— Jeff Dutky


 

Mark,

I had no luck uncrimping the can with a pair of dikes (assuming that means "diagonal cutters") and narrowly avoided bleeding all over my sacrificial specimen, but I am making slow progress by levering between the crimped lip and the circular insert (is that the "wafer?" It appears to be made of a hard plastic -- Bakelite?). I haven't gotten to the point where I can remove the plastic insert, but it's clear that I will get there with the patient application of brute force.

(UPDATE: I just discovered that the can I have selected as a sacrificial specimen has both a crimped can, but also metal ring that the can is crimped over, and retains the plastic disc that holds the contents in the can. This discovery has greatly accelerated the disassembly process. The sacrificial specimen is from a 7623A manufactured in the late 70s. Do the cans from the 50s have a similar construction? If so, I assume that you keep the ring for reassembly, as it has the outer ground pins attached to it)

You mention a black rubber-like substance, do you retain that for reassembly? I know that modern electrolytic capacitors have a rubber plug in the bottom that seals the contents in. I suppose that it's of no consequence here, because the capacitor we will stuff the can with already have their own rubber seals.

The can I have also has a plastic sleeve that appears to have been shrunk to fit he can. At least two of the big filter caps in the 533A have hard plastic cylinders over them, so I'm expected that they won't have the shrink tubing. If they do, however, I assume that has to be removed before you heat the can over the stove. Also if so, how do you replace the plastic sleeve?

Dan G,

Rest assured that I will be testing all components before replacing them, capacitors included. I don't have a working curve tracer (sadly) but I do have a DE-5000 LCR meter, whose limits I do not fully understand (need to read the manual). I have used the DE-5000 to test some expected bad capacitors from more recent vintage instruments and they have been correctly identified as bad, so I will likely rely on that to check the caps in the 533A.

As I said, this is not going to be more of a shelf queen than a daily driver, so I'd like it to look as original as possible, inside and out.

(that said, I've got this half-formed plan to add a feature to the instrument and it's scope-mobile cart that would allow me to have two extra plug-ins active in the cart. That would require some modification to the instrument, in the form of an external expansion interface, a power supply in the cart, and multiplexing circuitry in both the scope and the cart, but I intend to keep the defacement of the instrument to an absolute minimum, limited to cutting a hole in the rear of the case for a round, military-style Amphenol connector, mounting an entirely solid-state multiplexing board under the plug-in bay, and rerouting the plug-in interface wires through the multiplexing board)

-- Jeff Dutky


 

Mark,

Just a follow up: I understand much better now how you use a pair of dikes to open these cans: there's just a very small lip (~ 1mm wide) that holds the ring with the ground terminals in, and you only need to uncrimp that lip. I thought that the crimped lip was the full 4 mm wide ring holding the plastic disk in place. I also found out that I can just order a half-dozen meters of large diameter (50, 60, 80 mm, or more) clear heat shrink tubing, so I don't have any worries about replacing that now.

I am taking some pain to keep track of the orientation of the retaining ring (plus ground terminals) with respect to the can, the "wafer" and the positive terminal in the wafer. Some of this, I am sure, is unimportant, but certainly the arrangement of the ground terminals and the positive terminal matter, at least if I want to fit this can back into the circuit board it originally came out of.

As for heating the can to loosen the contents, I'm going to try using my hot air solder station instead of the kitchen stove, if for no other reason than to preserve some modicum of marital harmony.

It's amazing how much more sense even the best directions make once you get your hands dirty actually trying to follow them.

Thanks again.

-- Jeff Dutky


Mark Vincent
 

Jeff,

A small flat blade screwdriver may be needed to start the lifting for restuffing. That is between the ring and crimped over case on the bottom. The dikes are used to lift the outer case end up and go between the ring and outer case to lift it up. I do not have any I am restuffing at this time. Photos would be obvious to some things that words would take a long time to type. You are correct about the ground ring is held by pressure of the small lip pressed on it.

The ground ring has the small tab bent that fits in the notch on the phenolic wafer. You will see these when you pull the ring off. The orientation of the positive lugs is the important part in places where the wires/components are connected to them. You do not want to alter the original mounting. An example is to look under your 500 series at the wires to the lugs on the + and ring. A small can with the three negative lugs will need the wafer to be put back in the way it came out. Flipping it alters the connections 60 deg. off of original. The recent discussions of recapping 465/475/76x3 series shows why orientation is necessary. Usually the shape before the value on the outer can is in the schematic to which it is. There are times where a different voltage can be used, e.g. 160-200V condenser for a 100V supply instead of the 350-475V that was original. This way a higher capacitance can be used for better decoupling and the voltage is rationally close to the rated value of the condenser. The outer can is the one I prefer to turn in the way the characters are facing out when the unit is open for some reason to ease seeing when mounted. On a PC board condensers with radial leads or the pins, you cannot turn the condenser.

The heating is not a problem. The heat source at the top of the can with the + lugs facing up gets the tar gets warm enough to be soft. You will know it is ready to have the guts removed when holding the can using a mitt/rag/etc. with one hand and the other with something to pull on the + lug(s). Do not pull hard! You may need to pull on one lug then another to ease the guts out some then complete the removal at the bench. No need to completely remove the guts until it is at the bench. The mess will be at a place you can clean up after the work. Totally dry guts can be easy or hard to remove. There is no way to tell which it will be beforehand. Ones that are not totally bad usually are easy to remove. You can feel the guts to determine the condition. Dry ones (ones with high ESR) will be dry to the touch.

Ones I cannot restuff are ones that are the old wet types that use the large nut on the bottom with the + lug at the bottom and the case is - or the - is another lug. These are in early items from the late 20's to 30's and military or some other equipment. The early ones will have the top domed with the crimp at the top. The others will be flat on top. Another style that cannot be restuffed are ones where they are a can with the bottom being some fiberglass type with the + lugs riveted through the fiberglass. The fiberglass looking material is reddish, usually. This type is common in Japanese made items from the 60's and early 70's. These have to use a clamp to hold them as the - is a lug on the bottom with black paint on the rivet. I keep dud cans for replacement for the latter types in this paragraph. The large nut styles, I unsolder one lug, for a two lug style, with one lead on the lug that is still soldered to and the other end if bent out with the other end soldered to. I prefer to unsolder the + end and solder the wire(s)/component(s) to the + wire on the radial replacement. The - is soldered to - lug. It looks unprofessional yet is the only way I have found to do the job. You will need to determine the best way to do the job depending upon a case by case basis. Military types can have axial types with lugs for the connections. Those I put in a terminal strip where the clip mounting screw was to hold the condenser. When you see this type, you will know what I mean.

When I resfuff another can, I will get photos and post them here showing the stages of what I do. The can(s) may be in something other than the group's specific brand theme. The principle is the same no matter what device it is in. Because of the work restuffing, I will use the low ESR, high temp. and long life types to maximize the time before it needs to be restuffed again.

When I do not restuff a section of whole can is when it is something like a cathode bypass in audio amps in the low level stages, e.g. can with one or more 25mfd 25V in the can or vertical linearly section in a TV. Those I unsolder the lead at the pin(s) and solder the new condenser from the pin to ground or where the - goes.

Cans with the heatshrink over the can with a disc on the top cannot be restuffed while keeping the original heatshrink on it or putting it back on when finished. An example of these are in 465/475 series and others. Those I put in the new types and leave it. Ignore what the new ones look like inside. It will work when finished. If a new condenser is made at or larger diameter than a can, you must mount it as best possible. I have done this when I had to. You cannot always keep everything exactly original looking. I would rather have something be off original looking some and work than original looking and not work or not be right. This is a case by case basis.

The odd style is a even smaller can with two ground lugs and a single + lug. The place I have seen this one is in a H/P 5245L power supply for the nixie display B+. Those I put in a higher capacitance at 250V. These are harder to deal with being small. Gutting is easier being smaller. If you do one, you will know what I mean. They are unusual to see because few things used them.

The screw types in the 485 and 7000 series do not need heating to restuff them. The process of mounting the new condenser in them is different. The tar in these types comes out easily at room temperature. The material between the guts and tar makes removal of the guts easy, usually. Sometimes the "hard" tar will need to be broken out in pieces far enough to remove the guts then the remaining tar can be broken away and discarded leaving a clean can. There will be some white electrolyte to clean from the black end on the inside part. If requested, I will tell how to compete this process.

Getting your hands/arms dirty working on older electronics is expected. Ask your mechanic about dirty hands. Getting an old TV guts ready to be cleaned and restored will likely get your arms dirty. I have had this many times. Soap and water after the work is done. It is a good feeling to get something working again no matter the hassle. The satisfaction is worth the effort.

It will be great to hear when you get your 533A working the way you want. My restuffing procedure will likely be done by others on this group. I hope they get their piece restored and happy with a working piece.

Mark


Ed Breya
 

Jeff Dutky wrote: "I understand much better now how you use a pair of dikes to open these cans,,,"

In case it's not obvious, it's worth mentioning that you need to use flush-cut style dikes so you can grab the curled edge for lifting. Beveled edge dikes are no good for this. You dig in right on the surface of the steel ring, undercutting the lip and lifting at the same time. Once you get some peeled up, the rest becomes easier.

I too have described my methods over the years. I agree with Mark that the un-crimp/re-crimp methods are best, in situations where the caps can be safely removed and replaced. In cases where the caps are impractical to remove, there have been all sorts of suggestions on cutting them in place and leaving the bases attached to the circuits and such.

I've never had to resort to cutting cans in place, and have always un-crimped. My methods differ from Mark's, but the goal is the same - safely get the guts out, replace with new caps, and make it look good. Here's a summary:

For best appearance, and for caps that have plastic sleeving outside, you can protect the base rim of the can with some layers of tape, then an adjustable hose clamp around it. It is tightened snugly, but not so tight that it deforms the can. It can be placed so that the very bottom of the can rim is flush with the clamp edge, or possibly recessed slightly, up to about 1-2 mm, depending on the situation. Then the inner dike jaw digs into the lip, and the outer jaw levers against the clamp instead of nicking the soft outside of the base. The lip itself, and any plastic over-wrap will be unavoidably mangled in the process. The plastic may be trimmed off afterward, or even before the operation. To get a clean cut, the clamp can also be used as a cutting guide, when set to an appropriate position. The main thing is to preserve as much of the aluminum lip material as possible, and shape it back to an open cylinder as it was during its original construction, ready to crimp back together.

Once the can's base is nicely opened up, pry out the steel ring and phenolic base, by grabbing the terminal(s) with the dikes (not to cut of course, just grab - or use pliers if needed). Usually a good hand grip around the base, and prying against the thumb or fingers will give enough controlled force to extract the phenolic terminal base. Don't pry against the newly formed can lip material, or it will get mangled again. If necessary, the clamp can be re-positioned to protect it, and used for prying. The thin aluminum contact strips inside easily tear apart (or are often disintegrated away anyway - a common failure mechanism) from the phenolic, exposing the rubbery seal layers, that can be removed by poking and grabbing and prying with various tools and methods. Try to preserve all the pieces intact, and note the assembly order. There is some risk that the phenolic may be stuck in there too tightly, and it may break if the terminals are pulled too hard. Use judgement in applying force. It can be loosened up by moving the parts around relative to each other, with solvents, or heat. Also be sure that the can base is not deformed, which could pinch and bind it.

Once the cap slug is exposed, you can assess the situation. Most of the ones I've encountered were easy. If there's decent air clearance between the slug and the inside of the can, just jam the tip of a large flat screwdriver into the center of the slug - about half-way deep is enough - and tip the thing around and back and forth, and rotate. It will usually snap the slug loose from the glob of brittle tar at the top of the can, and it will pull out like a hotdog on a stick. If the slug is hard, the screwdriver may be hammered into it - again, not too deep. If the slug is a tight fit in the can, or the space is all filled with tar or other material, then it's tougher to pop out. In this case, you can partly bore out the slug with a large drill, being careful to go as deep as possible, but not so much that you accidentally go through the top of the can. With a big enough bore-out, you can collapse the slug layers inward for removal. You can use heat too, but I think boring it out is most expedient.

The cleanup phase depends on the materials, and how much is left. First try chipping it out - the old brittle tar often breaks into pieces. If it's still pliable and sticky, put it in the freezer for a while, then see if it will crack. If not, then scraping, heat and solvents will do. Also, be sure to thoroughly clean all the debris and juices off the tools.

Ed


Ed Breya
 

Forgot to mention that if you opt for using a hose clamp, you may have to re-position it at times, depending on how the clamp is built, to get around its own mechanism. Another option is to use an actual capacitor vertical chassis mount clamp (seldom used with twist-lock caps of course), I have a number of these, and had planned to try out the next time I do any re-stuffing. The problem there is getting close enough in size.

Ed


 

Mark, Ed,

Thanks for the additional instructions and clarification. I was, in fact, not using a flush cutter, but instead some cheap Chinese dikes that I was not worried about damaging. I have ordered a pair of cheap flush cutters, and if they work I will order a 10-pack that significantly reduces the per-unit cost, so I can use them for this kind of work without worrying about damaging them.

I have tested the resistance of the power supply rails to ground (except for +75 V and +325 V UNREG, which I could not find the test points for). Here are the results:

-150 V, 3 kΩ expected, 4.5 kΩ measured
+100 V, 500 Ω expected, 13.3 kΩ measured
+225 V, 6 kΩ expected, 18 kΩ measured
+350 V, 20 kΩ expected, 50 kΩ (37.9 kΩ) measured (reversed leads)
+500 V, 30 kΩ expected, 50 kΩ measured
T601 pins 1&4, ∞ expected, ∞ measured

All of these values (with the exception of the transformer pins) are higher than expected. What does that mean? I checked the values of some random AB carbon comp resistors, and with the exception of one in the HV supply, they all seemed to be in spec (surprisingly). I would expect leaky capacitors to be parallel paths to ground, and therefore REDUCE the resistance to ground, not increase it. Shorted capacitors even more so.

I'm holding off on my practice of unstuffing the capacitor until my roll of shrink tubing arrives (Amazon says it will be here tomorrow), then I will cut away the existing plastic sheath, and heat the can with my hot air pencil to loosen the contents. We will see how that goes.

-- Jeff Dutky