Topics

2445A slow

Steve
 

Well, I got my 2445A from a guy on eBay who said it had a slow startup which it did sort of I guess the first time I started one of these up? I have no idea if they’re instant on or I presume they have a startup routine to perform necessary system checks. After the first start which was slow compared to restarting it warm could definitely be a capacitance issue in the LV power supply. What do you guys mostly use to recap a supply, Sprague, vishay? I imagine tight tolerances. Also I tried calibrating a probe I tried 200 and 100 MHz probes as I’m still waiting on the correct probes to arrive and it doesn’t give any discernible pattern of that of a squar wave pattern it’s more like German army helmets during ww1 and my curve tracers forget it, it only went to xy mode but it didn’t display transistor characteristics hopefully I’m doing something wrong because this thing surely couldn’t be that broken.
--
Kind Regards,

steve

Chuck Harris
 

Hi Steve,

Understand that probe calibration is supposed to be done at 1KHz,
as it is intended to calibrate the DC to AC transition of the probe.

The 2445A's calibrator will change its frequency with sweep speed.
At the higher sweep speeds it gets positively ugly.

The only thing really all that special about the power supply caps
is they should be high temperature (105C) and long life. It is a
switching power supply, so Low ESR and high ripple current capacitors
make for better filtration and longer life.

I like to use higher voltage, and higher value capacitors in some
positions, selecting replacements that are physically about the same
size as the originals. I use only United Chemicon, Nichicon, or
Panasonic, and only from Digikey or Mouser. Never from amazon or
ebay.

I replace both the 180uf, 40V, and the 250uf, 20V with 330uf, 50V.
I replace the 100uf, 25V, and 100uf 35V with 100uf 50V. The 3.3uf,
350V gets a 3.3uf 350V, and the 10uf, 100V gets a 10uf, 100V.

The 1uf non polar can be replaced with like, or with ceramic,
capacitors.

It is rare that the two big axial leaded caps that filter the power
line voltage straight from the bridge/voltage doubler need replacement.
You won't be able to find an exact replacement. Go for something
that is the same physical size, but higher capacitance, in a radial
leaded package. You can hide the strange wiring necessary inside
of the black plastic covers with the capacitors. A little hot glue,
or craft (duco) glue can be helpful. Don't expect the capacitor's
plastic sleeve to insulate against the positive leads voltage. Use
sleeving.

The WIMA safety capacitors should be replaced with something similar
sized in a plastic dielectric.

The line input filter doesn't usually cause any problems in 120V
land, but might in 240V land. Your mileage may vary.

You might want to change the electrolytic capacitor(s) in the anode
supply board. High ESR on this part makes for noise everywhere.

Don't replace the electrolytic caps on the A1 main board, it is too
hard for very little gain. You are very likely to damage connectors
and cables the first time you try to remove the A1 board.

-Chuck Harris

Steve wrote:

Well, I got my 2445A from a guy on eBay who said it had a slow startup which it did sort of I guess the first time I started one of these up? I have no idea if they’re instant on or I presume they have a startup routine to perform necessary system checks. After the first start which was slow compared to restarting it warm could definitely be a capacitance issue in the LV power supply. What do you guys mostly use to recap a supply, Sprague, vishay? I imagine tight tolerances. Also I tried calibrating a probe I tried 200 and 100 MHz probes as I’m still waiting on the correct probes to arrive and it doesn’t give any discernible pattern of that of a squar wave pattern it’s more like German army helmets during ww1 and my curve tracers forget it, it only went to xy mode but it didn’t display transistor characteristics hopefully I’m doing something wrong because this thing surely couldn’t be that broken.

toby@...
 

On 2020-04-02 5:51 PM, Chuck Harris wrote:
Hi Steve,
...
You might want to change the electrolytic capacitor(s) in the anode
supply board. High ESR on this part makes for noise everywhere.

Hi Chuck

Noob question here. After doing many recappings on LCD monitors and
other simple inverters, I have finally got an ESR meter.

But I've never actually known what values to look for. How can we decide
whether a measured ESR is appropriate or not for power supply caps?
(Same as datasheet ESR when ordering for that matter.)

Thanks
--Toby


Don't replace the electrolytic caps on the A1 main board, it is too
hard for very little gain. You are very likely to damage connectors
and cables the first time you try to remove the A1 board.

-Chuck Harris

Steve wrote:
Well, I got my 2445A from a guy on eBay who said it had a slow startup which it did sort of I guess the first time I started one of these up? I have no idea if they’re instant on or I presume they have a startup routine to perform necessary system checks. After the first start which was slow compared to restarting it warm could definitely be a capacitance issue in the LV power supply. What do you guys mostly use to recap a supply, Sprague, vishay? I imagine tight tolerances. Also I tried calibrating a probe I tried 200 and 100 MHz probes as I’m still waiting on the correct probes to arrive and it doesn’t give any discernible pattern of that of a squar wave pattern it’s more like German army helmets during ww1 and my curve tracers forget it, it only went to xy mode but it didn’t display transistor characteristics hopefully I’m doing something wrong because this thing surely couldn’t be that broken.

Steve
 

Thank you for your reply Chuck, I’m a vintage hifi tech and I don’t really work on instruments in this magnitude simply because it’s not my forte. However I mostly use either nichicon upw series or vishay for axial and of course all are long life hi temp as you already know. I cleaned it up a bit on the inside and placed the cal selector to the cal position to run diagnostics and to the best of my ability it doesn’t seem to be throwing any codes in diagnostics and it went through it’s self calibration without any hesitation.

--
Kind Regards,

steve

Steve
 

I guess what I’m getting at is, is it urgent I do a recap and get the power supplies stable and at proper voltage and everything BEFORE I start hooking up my curve tracer and having not recapped it right away do you think it’s a ticking time bomb or am I safe for an unspecified time as I really need to use it and I don’t necessarily have time right now to do it. After removing the cover with the connectors I did a thorough visual check of the low voltage supply but I didn’t notice anything that is a problem like swollen caps or corroded leads but I know they’ve definitely seen there better days.
--
Kind Regards,

steve

Chuck Harris
 

The electrolytic capacitors in the power supply of 2465 family
scopes will not do any lasting harm if they are allowed to leak.
They tend to leak in two spots: Around the leads seals, and
through the aluminum sides (under the plastic sleeve).

The SMD electrolytic capacitors on a late model 2465B's A5 board
will do rapid, and often hard to repair damage if allowed to leak.

Your "A" scope won't have the SMD problem, nor will any plain model
scope.

It is an easy job replacing all of the electrolytic capacitors
in the power supply. It takes me no more than 30 minutes from
start to finish. I do it a lot, so expect no more than an hour.

-Chuck Harris

Steve wrote:


I guess what I’m getting at is, is it urgent I do a recap and get the power supplies stable and at proper voltage and everything BEFORE I start hooking up my curve tracer and having not recapped it right away do you think it’s a ticking time bomb or am I safe for an unspecified time as I really need to use it and I don’t necessarily have time right now to do it. After removing the cover with the connectors I did a thorough visual check of the low voltage supply but I didn’t notice anything that is a problem like swollen caps or corroded leads but I know they’ve definitely seen there better days.

Chuck Harris
 

Hi Toby,

Any ESR meter you are likely to own is simply a 50KHz
AC ohmmeter.

Every capacitor has both its intrinsic capacitance,
and an effective series resistance, so it looks like this:

-------ESR-----CAP------

The ESR meter works in a range of frequencies where the
capacitor's capacitive reactance is much lower than the
normal value for ESR.

Capacitive reactance is:

Xc = 1/(2*pi*f*C) Where f is in Hz, and C in Farads.

You can rearrange the equation, and plug in some values,
and find that at 50KHz, capacitors Xc starts to dominate
the "resistance" around 1uf... 1uf has an Xc of about
3 ohms at 50KHz.

Here are a few more Xc's (at 50KHz):

0.1uf -> 30ohms
1uf -> 3ohms
10uf -> 0.3ohms
100uf -> 0.03ohms

Remember those relationships, and life becomes much simpler.

You should expect the ESR to be no lower than the Xc
of the capacitance you are measuring (see above table).

But, you also should expect it to be only minimally higher.

If you measure a 10uf cap, and it is 10 ohms, you have
a problem. If it is 1 ohm, probably not... depends on
whether it is being used as a filter cap, or an audio
capacitor.

Typically, when a capacitor has died to ESR, you will find
that its measured ESR is in the whole ohms to infinite range.

A good capacitor will typically be less than one ohm ESR.

-Chuck Harris

toby@... wrote:

On 2020-04-02 5:51 PM, Chuck Harris wrote:
Hi Steve,
...
You might want to change the electrolytic capacitor(s) in the anode
supply board. High ESR on this part makes for noise everywhere.

Hi Chuck

Noob question here. After doing many recappings on LCD monitors and
other simple inverters, I have finally got an ESR meter.

But I've never actually known what values to look for. How can we decide
whether a measured ESR is appropriate or not for power supply caps?
(Same as datasheet ESR when ordering for that matter.)

Thanks
--Toby


Don't replace the electrolytic caps on the A1 main board, it is too
hard for very little gain. You are very likely to damage connectors
and cables the first time you try to remove the A1 board.

-Chuck Harris

Steve wrote:
Well, I got my 2445A from a guy on eBay who said it had a slow startup which it did sort of I guess the first time I started one of these up? I have no idea if they’re instant on or I presume they have a startup routine to perform necessary system checks. After the first start which was slow compared to restarting it warm could definitely be a capacitance issue in the LV power supply. What do you guys mostly use to recap a supply, Sprague, vishay? I imagine tight tolerances. Also I tried calibrating a probe I tried 200 and 100 MHz probes as I’m still waiting on the correct probes to arrive and it doesn’t give any discernible pattern of that of a squar wave pattern it’s more like German army helmets during ww1 and my curve tracers forget it, it only went to xy mode but it didn’t display transistor characteristics hopefully I’m doing something wrong because this thing surely couldn’t be that broken.




tekscopegroup@...
 

A good start would be to check the all the power supplies voltage and ripple content, according to the service manual. That should give you an idea of how urgent recapping attention is needed.

Chuck Harris
 

Unfortunately, it doesn't always work that way.
The caps tek used leak through the lead seals long
before they cease being good enough to meet the
ripple specs.

I think the capacitor manufacturer knew that leakage
was going to be a problem, because in addition to the
usual rubber lead seal, they filled the area after the
seal with red, or black epoxy. The leaks are where
the leads pass through the epoxy, and show up as some
black crust around each lead... that usually also ends
up on the board itself.

It is always a good idea to check voltage and ripple,
but recapping the supply is pretty easy and cheap to
do.

From working on a lot of these supplies, the cap on
the -15V supply's inverter is almost always bad.

-Chuck Harris

tekscopegroup@... wrote:

A good start would be to check the all the power supplies voltage and ripple content, according to the service manual. That should give you an idea of how urgent recapping attention is needed.



Brad Thompson
 

Chuck Harris wrote on 4/3/2020 10:48 AM:

Unfortunately, it doesn't always work that way.
The caps tek used leak through the lead seals long
before they cease being good enough to meet the
ripple specs.

I think the capacitor manufacturer knew that leakage
was going to be a problem, because in addition to the
usual rubber lead seal, they filled the area after the
seal with red, or black epoxy.
<snip>

Hello--

IIRC, at one of my former employer's facility, the solder flux-remover solvent in use at the time attacked
conventional electrolytics' rubber end seals. The epoxy end seals prevented the damage
but those capacitors were more expensive. Again IIRC, the PC-board manufacturing process changed to
a different flux and solvent that didn't attack the rubber end seals.

73--

Brad  AA1IP

Chuck Harris
 

I have heard that, and no doubt it was at one point
true.

As I recall, Mallory and Sprague advertised the epoxy
sealed capacitors as being ultra reliable, or some such
thing.

If the cleaning solvents were the ultimate problem with
the rubber seals, then the epoxy should have protected
the rubber from the solvents and other environmental
concerns... 30 years is not bad.

-Chuck Harris

Brad Thompson wrote:

Hello--

IIRC, at one of my former employer's facility, the solder flux-remover solvent in use
at the time attacked
conventional electrolytics' rubber end seals. The epoxy end seals prevented the damage
but those capacitors were more expensive. Again IIRC, the PC-board manufacturing
process changed to
a different flux and solvent that didn't attack the rubber end seals.

73--

Brad  AA1IP

Steve
 

Thank you for your info on that, I looked but I didn’t see any leakage but I also know that doesn’t mean anything. I will go ahead and recap the LV supply if it’s an hour job, I can afford an hour. I will recap the LV supply according to Chuck’s message.
--
Kind Regards,

steve

Chuck Harris
 

To avoid mistakes know a couple of things:

1) all capacitors have their "+" leads pointing in the
same direction.
2) only remove one *type* of capacitor at a time, and
replace it before you move on. There are marking errors
on some versions of the supply that can cause you to swap
two capacitors to one capacitor's peril.

-Chuck Harris

Steve wrote:


Thank you for your info on that, I looked but I didn’t see any leakage but I also know that doesn’t mean anything. I will go ahead and recap the LV supply if it’s an hour job, I can afford an hour. I will recap the LV supply according to Chuck’s message.

Steve
 

Thank you for the tip chuck! I sincerely appreciate all the help. If I can ask one last question or two. I know you upped a few caps to 330uf which while upping voltage is ok upping value generally speaking is not. I’m just curious for the reasoning in increasing the value. And ironically I have on hand Nichicon uvz 3.3uf 350v as I had to recap a pioneer reel to reel machine. I had purchased twenty and thought I’d never use another one of those caps. Boy was I wrong, glad I have them now.

--
Kind Regards,

steve

Steve
 

Also another thing I meant to ask, I’m neither very familiar with this scope or an engineer so the service manual can become quite daunting. I’m still trying to get my curve tracers to work properly when I switch to xy but all I’m getting are two angled lines without the hfe comb and the service manual hasn’t been much help. It’s like I need an interpreter in some instances.

--
Kind Regards,

steve

Steve
 

You’ve also mentioned the wima safety caps but I’m not seeing any of those. And you had stated in one message 2465 and I just wanted to make certain that before I follow your cap change instructions that this applies to the 2445a?
--
Kind Regards,

steve

Chuck Harris
 

Hi Steve,

The capacitors that I upped in value are the six that are most often
damaged due to internal heating. They are the capacitor filters
for the inverter's +15V, +5V, -5V, -15V Unregulated, and the +5.1VD
regulated supplies.

The original capacitors were 180uf40V +100%-10%, and 250uf20V +100%-10%
capacitors. The new capacitors, are 330uf, +/-10%.... at least with
the brands I specify (United Chemicon, Nichicon, and Panasonic). I have
measured lots of them, and they are more like +/-5% in my experience.

180uf +100% = 360uf vs 330uf +10% = 363uf
250uf +100% = 500uf vs 330uf +10% = 363uf

So, the replacement capacitors are very close to being within the
original tolerance.

The additional capacitance noticeably reduces the high frequency ripple
from the inverter.

Modern capacitors are all built to barely meet their published
specifications with the absolutely minimum cost and volume... the bean
counters like it that way. If you buy an new exact valued capacitor,
you will find the new capacitor to be about 1/4 to 1/2 the size of the
original. It will get hotter at as its only way to eliminate the heat
due to ESR losses is to radiate it. Radiation efficiency is directly
related to the surface area of the capacitor's case, which is related
to its volume.

Raising the capacitance lowers the ESR, and as such lowers the internal
heating in a given circuit.

Raising the capacitor's voltage rating lowers the ESR, and as such
lowers the internal heating in a given circuit.

Raising both the capacitance, and the voltage gives a capacitor that
more readily fits the board layout, putting less stress on the already
weak rubber lead seals.

Making all six capacitors the same type reduces my inventory, and
makes a recapping job quicker, easier, and less prone to mistakes.

I have done this change for at least 10 years, and have no failures.

-Chuck Harris


Steve wrote:

Thank you for the tip chuck! I sincerely appreciate all the help. If I can ask one last question or two. I know you upped a few caps to 330uf which while upping voltage is ok upping value generally speaking is not. I’m just curious for the reasoning in increasing the value. And ironically I have on hand Nichicon uvz 3.3uf 350v as I had to recap a pioneer reel to reel machine. I had purchased twenty and thought I’d never use another one of those caps. Boy was I wrong, glad I have them now.

Chuck Harris
 

All of the power supplies in the 2465 family are the same +/- the
fan mounting hardware.

The safety capacitors are yellow, and will have a clear epoxy coating
that makes the outer foil visible. They are paper, and the epoxy will
be cracked.

The don't need to be safety capacitors, as they are not directly across
the power line, but rather are being used as snubbers, with a series
flame proof resistor. They are 0.068uf, 330VAC rated, and are located
on the inverter board.

There are a couple of other in smaller values, that are even less likely
to be a problem.

-Chuck Harris

Steve wrote:


You’ve also mentioned the wima safety caps but I’m not seeing any of those. And you had stated in one message 2465 and I just wanted to make certain that before I follow your cap change instructions that this applies to the 2445a?

Chuck Harris
 

The 2465 family's scopes have a row of buttons that
enable or disable the various channels for display.

In the X-Y mode, the CH1 input becomes the X axis, and
the CH2 input becomes the Y axis.

Most people look at that and reason I am using CH1, and
CH2, so I must push both buttons to enable both axes.

The 2465 doesn't exactly work that way. If you push
both sets of buttons, you will get a trace for CH1 that
is angled, and a trace for CH2 that is straight.

Try it again, but don't press any of the buttons.

-Chuck Harris

Steve wrote:

Also another thing I meant to ask, I’m neither very familiar with this scope or an engineer so the service manual can become quite daunting. I’m still trying to get my curve tracers to work properly when I switch to xy but all I’m getting are two angled lines without the hfe comb and the service manual hasn’t been much help. It’s like I need an interpreter in some instances.

n4buq
 

I know my 2445 has the snubbers. In an effort to find what had smoked, I decided to shotgun all the larger electrolytics as well as the the snubbers. After replacing the electrolytics (and, unfortunately, finding them all testing good), the last thing I replaced was the snubbers and it was only then that I noticed that one of them had blown through a crack on one end. Wish I'd seen that one first but maybe I bought some extra time with new electrolytics on those boards.

Thanks,
Barry - N4BUQ

----- Original Message -----
From: "Chuck Harris" <cfharris@...>
To: TekScopes@groups.io
Sent: Saturday, April 4, 2020 6:52:08 AM
Subject: Re: [TekScopes] 2445A slow

All of the power supplies in the 2465 family are the same +/- the
fan mounting hardware.

The safety capacitors are yellow, and will have a clear epoxy coating
that makes the outer foil visible. They are paper, and the epoxy will
be cracked.

The don't need to be safety capacitors, as they are not directly across
the power line, but rather are being used as snubbers, with a series
flame proof resistor. They are 0.068uf, 330VAC rated, and are located
on the inverter board.

There are a couple of other in smaller values, that are even less likely
to be a problem.

-Chuck Harris

Steve wrote:

You’ve also mentioned the wima safety caps but I’m not seeing any of those.
And you had stated in one message 2465 and I just wanted to make certain
that before I follow your cap change instructions that this applies to the
2445a?