Bruce, the + sign is unambiguous, that lead is definitely the postive lead.? The 'bent knee' lead may be the + lead but I am not aware of that as a convention so trust the markings.? Likewise the square solder pad may be a Tek convention (can other readers comment?) but I am not aware of it so I wouldn't rely on it.? Not having access to a 475 I can't get a good look at the A7 board and the image of the PCB layout, Fig 7-16 in the manual is very grainy and I can't see the shape of these pads.? The best way to confirm which way around these tantalums should be is by looking at the pics you took of the board before you removed them!? Yes I forget to do that too, but it's very helpful in these circumstances.? I did refer to the pic of C1091 you posted but the focus is a bit off where it's needed.? I *think* the + is on the left as that looks like the 'bent knee' lead but I'm not certain.
The next best way is to identify nearby components and find which connects to which, and the schematic shows those two ferrite bead inductors L1091 and L1093 and they tell the story.? Note that supply rail to L1091 and C1091 is negative and the other L and C are on a postive supply rail so the capacitor polarity will reflect that.
As for the colourful tantalum on the A9 board, I worked out the colour code as you did and I can't explain the difference with the parts list.? That could be a wild goose chase all of its own so unless you have reason so suspect its motives, I think the sleeping dog principle applies.
Graham
toggle quoted message
Show quoted text
On 28/05/2020 9:43 am, ciclista41 via groups.io wrote: Hi Graham,
I try to treat people that way, too. After my brief adventure with C1091 and C1093, I'm doing the same with tantalum caps in this scope.
The leads on this tantalum are not symmetrical to the body. If you look at them as legs on a body, one seems to have a knee bent out to the side rather than dropping straight to the floor. There is a + sign printed closer to that leg. I'm guessing that's the positive lead. The through holes they came out of have a square trace pad and a round trace pad. I'm guessing these are to guide polarity. I think the positive leg went to the square, while the negative went to the round, but I'm not absolutely sure of that. Confirmation? Is this a standardized thing?
At /g/TekScopes/photo/247625/0?p=Created,,,20,2,0,0, you'll see an even more colorful one I found on my A9 board. Who knew these scopes were wearing jewelry when no one was looking?
According to this chart: , that's a Red=2, Violet=7, Grey=.01 multiplier, Green = 16V.
So that's .27μF, right? But the manual says this is a 196D275X9050JA1 or a 290-0573-00. The latter is listed in the Tektronix Common Parts Catalog as 2.7μF, 50V. What am I not getting?
Bruce
On Wed, May 27, 2020 at 07:40 AM, VK1GVC wrote:
Bruce, I treat tantalums like I treat people - assume they have good
intentions unless I learn otherwise.? If you remove lots of them without
reasonable suspicion then you'll probably do more damage than good as
Eric has warned.? As tantalums usually fail to a short circuit, a
failed/shorted tantalum will pull down the voltage at that point to
close to zero and this will/should be straighforward to find - later.
At this stage, if I haven't missed any key points, you've found 2
sick/dead capacitors in the +50V supply which MAY be the cause of all
the wrong voltages you've measured in the low voltage power supply
section.? So replace those caps first, then adjust all of the LV power
supply to meet calibrations specs and see what happens.? You might have
a working scope if luck is very much with you, or you might be one step
on a long journey towards that goal.
And when you replace the tantalum which lost a leg, make sure you
install the new one with the right polarity.? Tantalums don't like
reverse voltage and they can get hot, emit smoke and explode.? I've seen
it happen when a workmate powered up some newly assembled eqpt for the
first time and heard a hissing sound so he bent down for a better look.
A cylindrical metal tube type tantalum then exploded and flew off the
PCB, burning itself onto the side of his neck. Very painful ... and the
language was terrible.? And very lucky he didn't get hit in an eye.
BTW here's a different type of tantalum which I don't think has been
mentioned yet - see the colourful little blob in the lower left of the
picture at this link:
/g/TekScopes/photo/12901/22?p=Name,,475,20,1,0,0
It has a black top, brown middle and green bottom ... the colours are a
derivative of the resistor colour code and I think there's a spot on one
side too.? Much better dressed than a plain boring tantalum!
Graham
On 27/05/2020 3:28 pm, ciclista41 via groups.io wrote:
Graham,
Well, my scope is peppered with the kind in the photo. Maybe they re-spec'ed the caps later in production, or maybe the US Army bought enough in
their contract with Tek that they were able to spec what they wanted. Mine is
from November of 1982, if the date code on the AC power cord box inside the
case just above the transformer is original.
I went ahead and pulled the two tantalums out of the circuit to measure them. I broke a lead off the first one, pulling too hard. Because I only
have access from the top of this A7 board unless I remove it, which I don't
want to do, my solder sucker wasn't able to slide over the clipped end of the
leads, and I had to just use a sharp tip on my iron. Did much better on the
second one, which came out easily by that second method. I then used the
sucker to remove the remaining solder from the holes in the board for when I
replace the broken one.
Yes, you were right to question my belief that they were bad. Out of circuit, they both tested well within the green on my ESR tester. So much for
being able to reliably test capacitors in circuit with that thing. They
weren't even close to being in spec when tested with it in circuit. I was
thinking that these dipped tantalums were generally bad in boards this old
based on numerous threads in this forum and others saying how often these were
bad. Some said they routinely replace them all when they see them in a scope
of this era. I have tested dozens on this board and only found about a handful
that this tester considered bad. So, considering that these two were among
the worst that I tested in circuit, I decided to pull them, as they'd be less
easily accessible later when the A8 board is back in place.
So this makes me question whether the tantalums in this board are among the infamous ones. I hope not, obviously. Your statement that my photo showed
what looked like a very '80's tantalum made me think maybe the infamous ones
are some earlier 60's or 70's version. How can I know, other than pulling
them all? At this point, I don't see the point of pulling any of them unless
they show up as faulty later, when there is a trace on the scope and I can run
it through its paces and make sure it's working as it should.
Speaking of working as it should, Tektronics calls for several specialized calibration tools in the manual. If the scope is running well, with no
serious issues other than calibration, can it be calibrated without resorting
to the purchase of such tools? I briefly noted some shops wanting a couple of
hundred dollars to calibrate a scope. Definitely don't want to do that! If I
thought I had to have someone else calibrate it, I probably wouldn't have
taken on the project and just invested in a new digital scope.
Bruce
On Tue, May 26, 2020 at 09:09 PM, VK1GVC wrote:
Bruce, the link to the pics on TekScopes worked for me and that
capacitor looks very 1980's tantalum as Michael confirmed in a later 475
manual and the Tek Common Parts manual.? Ceramics of that era were
commonly flat circular disks, very different from what you have.
Roger re testing in place - very problematic.? Best to remove them to
eliminate any ambiguity and as you have reported success in desoldering
then that's the best option while you have access.? BTW I'm curious to
know why you *believe* that they are faulty - mere suspicion based on
type and age, or something else?
You quoted your post of 22 May to Harvey a few minutes ago in which you
sought advice about what can and cannot be substituted when replacing
1982 components in 2020.? Has that qn been answered to your satisfaction
or is it still a live issue?? The short answer is: it depends.? The long
answer really has to address a specific component in a specific
application.? But the laws of physics haven't changed a lot in the last
40 years so there is almost certainly something out there which can be
bought/found/made/adapted or cajoled to do the job.? If you need a 1amp
400V rectifier diode then a IN4004 of the 70's or 80's is just the same
as one from the factory today.? If you need a very specific high-spec
module made for or by Tek for a very challenging application 40 years
ago and now out of production ... oh dear, you've got a problem.
Fortunately we now have TekScopes, some wikis, eBay and of course the WWW.
Graham
On 27/05/2020 1:24 pm, ciclista41 via groups.io wrote:
Hi Graham,
I posted a photo of the C1091 tantalum, along with the plaque stuck to the back of my scope here:
/g/TekScopes/album?id=247625&p=Created,,,20,2,0,0
Bruce
--
This email has been checked for viruses by Avast antivirus software.
|
