Topics

2247A PSU Troubleshooting

Nicholas Keller
 

Hello, first post here and first real scope. A tech friend of mine gave me this in broken condition and said I could keep it if I can fix it. So here goes....

It doesn’t boot. Nothing happens when I turn it on.

I have disassembled the unit and visually inspected the PSU board as he said it had been repaired once before and was likely the same fault. He can’t find the service record. I couldn’t tell that anything had been replaced, and couldn’t see any obviously bad parts (swollen, leaking caps, for example).

I reassembled and powered up to test the power headers on the underside of the unit. They are all basically dead, however I did measure some mV on one of the high voltage pins.

I have a PDF of the schematic. I saw mentioned in a previous post about a record of some common issues with this PSU, but I couldn’t find the old post detailing these. Can anyone give me some suggestions before I attempt to trace signals? Is it safe to remove the PSU board and apply power to it even though some connections appear to be via other boards?

Thanks so much,
Nick

 

On Mon, Oct 22, 2018 at 03:22 PM, <Nirokeforums@...> wrote:


It doesn’t boot. Nothing happens when I turn it on.



I reassembled and powered up to test the power headers on the underside of the unit. They are all basically dead, however I did measure some mV on one of the high voltage pins.
Did you check the mains fuse and switch?

Raymond

Nicholas Keller
 

Fuse: yes.

Switch: no, and I should have because that’s an easy one and have had bad
switches on other gear

Thanks, Raymond, I will check the switch tonight after work

Nick



On Mon, Oct 22, 2018 at 10:14 PM Raymond Domp Frank <@Raymond>
wrote:

On Mon, Oct 22, 2018 at 03:22 PM, <Nirokeforums@...> wrote:


It doesn’t boot. Nothing happens when I turn it on.



I reassembled and powered up to test the power headers on the underside
of the unit. They are all basically dead, however I did measure some mV on
one of the high voltage pins.
Did you check the mains fuse and switch?

Raymond



tekscopegroup@...
 

I have a 2247A myself, excellent GP scope.

One of the first recommended things to do I was told is replace all the rectifier diodes in the A18 LV power supply that have ZM or ZS markings on them as they are well known to get leaky after "some time" (quoting the very knowledgeable person that sent me this recommendation). Replace these diodes with MUR160 600V 1Amp, you easily get them at Mouser, Digikey, etc. On my scope all of them had either of the above suspect markings, so just to be safe I replaced all of them. Whole procedure takes maybe an hour or two, your mileage may vary.

Of course for an instrument this age, needless to mention to also check filter caps, specially if you see higher than normal ripple on any of the supply rails.

Suspect diodes are Tek PN 152-0400-00 (total 17 diodes) with alternate PN given as MB2501. These are the on-board part locator numbers:
-CR2202
-CR2204
-CR2205
-CR2208
-CR2209
-CR2210
-CR2211
-CR2212
-CR2213
-CR2214
-CR2215
-CR2216
-CR2218
-CR2227
-CR2228
-CR2235
-CR2236

Hope that helps.
Alex

Nicholas Keller
 

Thank you, Alex. This must have been the suggestion I saw here not too
long ago, I just got the impression that the diode list was an abridged
tally of what “should” be checked and replaced. When I pull the board
again, I will check these diode markings. Unfortunately I can’t test for
ripple yet as there is no power getting to the supply rails. I’ll report
back when I’m able to make some progress. Thanks again!

Nick

On Tue, Oct 23, 2018 at 8:09 PM <tekscopegroup@...> wrote:

I have a 2247A myself, excellent GP scope.

One of the first recommended things to do I was told is replace all the
rectifier diodes in the A18 LV power supply that have ZM or ZS markings on
them as they are well known to get leaky after "some time" (quoting the
very knowledgeable person that sent me this recommendation). Replace these
diodes with MUR160 600V 1Amp, you easily get them at Mouser, Digikey, etc.
On my scope all of them had either of the above suspect markings, so just
to be safe I replaced all of them. Whole procedure takes maybe an hour or
two, your mileage may vary.

Of course for an instrument this age, needless to mention to also check
filter caps, specially if you see higher than normal ripple on any of the
supply rails.

Suspect diodes are Tek PN 152-0400-00 (total 17 diodes) with alternate PN
given as MB2501. These are the on-board part locator numbers:
-CR2202
-CR2204
-CR2205
-CR2208
-CR2209
-CR2210
-CR2211
-CR2212
-CR2213
-CR2214
-CR2215
-CR2216
-CR2218
-CR2227
-CR2228
-CR2235
-CR2236

Hope that helps.
Alex



tekscopegroup@...
 

Hi Nick, can't really give you other troubleshooting tips for a dead 2247A power supply, as I fortunately never had this problem with my own scope, but I hope in the end its something easy to fix. Maybe one of the 22xx series experts here would have some useful tips for you. You mention that you have the schematic, not sure if you actually have the complete service manual as well. If not I would strongly suggest to download a copy which just in case is available on the right hand column on this page:

http://w140.com/tekwiki/wiki/2247

Good luck with the troubleshooting, and don't forget to (hopefully) keep it fun. Also, please report back any progress so the information might possibly be useful for the next person having to go through the same scenario as you are now.

Alex

nielsentelecom@sbcglobal.net
 

Nick,

I have a 2246A that I had to overhaul, which most likely has the identical power supply for your 2247A. I would do a search for the 2246A power supply also here. So I will refer to the component number in your schematic if there is a difference. I wouldn't be concerned with the diodes and caps mentioned in Alex's post until you verify the following.

I also blew up U2201 when I accidentally shorted my scope probe shield against a heatsink while probing. That caused a no power on the main 44VDC bus measured at C2203. even if that is good, I would get a spare replacement. they are not expensive and still available as of a couple years ago.

Get the service manual, refer to page 97, or 3-59, the power supply block diagram, and verify the voltage indicated prior to the preregulator and startup circuit at c2202. that is the main DC bus that feeds the Switching power supply pre-regulator. If you are plugged into 120VAC, it should be at about 160-170VDC. then switch meter to AC, and see if there is any ripple there. That's 120 hz so any meter will work. If that is good then check the previous paragraph for the 44VDC check.

I also was able to test my power supply out of the cabinet. It was risky, but necessary. I used a 10KV AC rated linemans glove and placed the CRT lead inside of it. I also have the resistor values that were close loads for the power supply to operate at a typical load.

NielsenTelecom

Nicholas Keller
 

I finally got a chance to take a look at this 2247a. At the test point you
mentioned (c2202, before the startup and pre-regulator, marked 14A on the
schematic), I’m only getting 77VDC, but got about the same 77V when I
tested for AC. I measured 120-122VAC up to the diode ring. Does this
suggest that C2202 is bad? Or something else?

Thanks!

Nick





On Thu, Oct 25, 2018 at 12:55 PM @Nielsentelecom <
@Nielsentelecom> wrote:

Nick,

I have a 2246A that I had to overhaul, which most likely has the
identical power supply for your 2247A. I would do a search for the 2246A
power supply also here. So I will refer to the component number in your
schematic if there is a difference. I wouldn't be concerned with the diodes
and caps mentioned in Alex's post until you verify the following.

I also blew up U2201 when I accidentally shorted my scope probe shield
against a heatsink while probing. That caused a no power on the main 44VDC
bus measured at C2203. even if that is good, I would get a spare
replacement. they are not expensive and still available as of a couple
years ago.

Get the service manual, refer to page 97, or 3-59, the power supply block
diagram, and verify the voltage indicated prior to the preregulator and
startup circuit at c2202. that is the main DC bus that feeds the Switching
power supply pre-regulator. If you are plugged into 120VAC, it should be at
about 160-170VDC. then switch meter to AC, and see if there is any ripple
there. That's 120 hz so any meter will work. If that is good then check the
previous paragraph for the 44VDC check.

I also was able to test my power supply out of the cabinet. It was risky,
but necessary. I used a 10KV AC rated linemans glove and placed the CRT
lead inside of it. I also have the resistor values that were close loads
for the power supply to operate at a typical load.

NielsenTelecom



 

On Mon, Mar 4, 2019 at 12:17 AM, Nicholas Keller wrote:


I measured 120-122VAC up to the diode ring. Does this
suggest that C2202 is bad?
I guess you realise you're measuring in an area that is directly connected to the mains, no isolation? Take great care! *Do* use an isolation transformer and *float your 2247A*.

Did you measure the 77 V across C2202 or between C2202 and chassis? The latter would be incorrect, because the chassis is floating (with exception of some C's) against C2202 (since that has a galvanic coupling to the mains).

Raymond

Nicholas Keller
 

No, I was not using an isolation transformer (I don’t have one) and yes I
was measuring between cap and chassis. I will look into getting a
transformer before testing across c2202 or doing other troubleshooting.

Thanks for the warning

Nick


On Sun, Mar 3, 2019 at 6:49 PM Raymond Domp Frank <@Raymond>
wrote:

On Mon, Mar 4, 2019 at 12:17 AM, Nicholas Keller wrote:


I measured 120-122VAC up to the diode ring. Does this
suggest that C2202 is bad?
I guess you realise you're measuring in an area that is directly connected
to the mains, no isolation? Take great care! *Do* use an isolation
transformer and *float your 2247A*.

Did you measure the 77 V across C2202 or between C2202 and chassis? The
latter would be incorrect, because the chassis is floating (with exception
of some C's) against C2202 (since that has a galvanic coupling to the
mains).

Raymond



Nicholas Keller
 

Raymond,

Per your advice (and that clearly marked on the PCB!), I have purchased an
isolation transformer. After watching a few videos online, I am still
unsure whether I should connect ground from the Tek to the transformer.
One video showed use of a 3-prong to 2-prong adapter where the ground wire
is left unattached. Is this what you describe as “floating the 2247A”?

Thanks for your help!

Nick



On Sun, Mar 3, 2019 at 8:48 PM Nicholas Keller via Groups.Io <Nirokeforums=
gmail.com@groups.io> wrote:

No, I was not using an isolation transformer (I don’t have one) and yes I
was measuring between cap and chassis. I will look into getting a
transformer before testing across c2202 or doing other troubleshooting.

Thanks for the warning

Nick


On Sun, Mar 3, 2019 at 6:49 PM Raymond Domp Frank <@Raymond>
wrote:

On Mon, Mar 4, 2019 at 12:17 AM, Nicholas Keller wrote:


I measured 120-122VAC up to the diode ring. Does this
suggest that C2202 is bad?
I guess you realise you're measuring in an area that is directly
connected
to the mains, no isolation? Take great care! *Do* use an isolation
transformer and *float your 2247A*.

Did you measure the 77 V across C2202 or between C2202 and chassis? The
latter would be incorrect, because the chassis is floating (with
exception
of some C's) against C2202 (since that has a galvanic coupling to the
mains).

Raymond





 

On Wed, Mar 13, 2019 at 03:17 PM, Nicholas Keller wrote:


Is this what you describe as “floating the 2247A”?
Hi Nicholas,
Glad you bought an isolation transformer!
My advice to "float your 2247A" actually was very badly formulated. I meant to say something like "float the hot (primary) circuits of your 2247A", meaning the circuits directly connected to the mains *and* I should have added *not* to float your testing equipment (like your testing 'scope).
Many people disconnect safety ground on their measuring 'scope but that's *very* bad and dangerous practice.

Always make sure all instrument chassis are connected to safety ground if at all possible.

Your 2247A is an example where you can get the protection of your isolation transformer and still connect the chassis of both your testing 'scope and the device under test (DUT) to safety ground. This is because T2204 takes care of isolation of all low-voltage circuits from the mains, *including* the 'scope's chassis, which is connected to low-voltage/ secondary ground.
The isolation transformer makes everything directly connected to the two mains input wires (the hot circuits) float i.e. galvanically disconnected from "the rest of the world".
Since the 2247A's chassis is isolated from the hot circuits, it's safe and actually a good idea (safe) to connect that chassis to safety earth. The ground connection on your isolation transformer always connects that transformer's housing to safety earth and the ground wire you're mentioning is supposed to connect that earth ground through to the 'scope's ground (3rd pin in 3-prong adapter).

If you connect things this way, your testing instruments will remain connected to ground, as will your 2247A's chassis.
Only the hot (primary) circuits of your 2247A will be isolated from mains via the isolation transformer on their input and from the 2247A's chassis through T2204 on their output.

Four final notes:

1. In most instruments, there are one or more capacitors between the mains connection(s) and chassis. These are there for EMC reasons but they may present a high-impedance voltage on the chassis (of Vmains/2) if safety ground is disconnected in that instrument.
2. Providing safety earth continuity to your DUT (where possible) provides protection in case of leaks or one of the capacitors as in 1. short-circuiting.
3. In most (older) instruments with linear supplies, like the 465 'scope, providing this earth continuity while using an isolation transformer won't do much good. In those instruments. By all means use an isolation transformer but skip the earth continuity.
4. Be aware that, though disconnected from mains, hazardous voltages still exist inside hot circuits connected via an isolation transformer!

Good luck with your repair!

Raymond

Nicholas Keller
 

I was simply using a hand held DMM (craftsman 82400) to test the voltage at
C2202. I do not own another bench scope. So if I understand correctly, I
can connect the DUT to the isolation transformer using all three prongs.

I will check the value at this capacitor this evening

Nick

 

On Wed, Mar 13, 2019 at 08:59 PM, Nicholas Keller wrote:


So if I understand correctly, I
can connect the DUT to the isolation transformer using all three prongs.
That is correct.

Raymond

chipbee40
 

The common faults and solutions to the power supplies on ALL the 22xx scopes are in the archives of this group. seriously this group goes over the same stuff repetitively :)
The 2246 (applies to the 2247) power supply can be removed from the scope and using the fan as a load can be tested as a unit.(With care! this is a mains potential switching supply).
Common failure on these is C2210 4.7uF and in many cases that's all, but if you are unlucky then Q2209/10 which if I remember correctly are TIP32C, following on in the unlucky stakes the mosfet Q2214.
The secondary diodes are rarely at fault, mostly schottky so reading low on diode ranges. and in my opinion the secondary 3 leg caps are reliable.
If testing the board watch out for a bite from J2208 long after you power off! (Near the multiplier, drain it with a 100k resistor). Rarely T2204 fails, also I've had the multiplier fail, (Disconnect its input to confirm).

Brian Symons
 

*Re: Isolation Transformers*

When using an isolation transformer for troubleshooting you do not want an earth being connected through.
The whole idea is to break any reference between earth & the mains.
The winding will make both outputs fully isolated from earth so that you can measure them safely.
This is one time that earth is your enemy.

Please physically check the item to make sure that there isn't an earth & test the output winding for insulation to earth.
Just in case, test both sides of the outlet (I know it should be continuous) & the higher the insulation reading (Megohms) the better.
If the transformer was being used for filtering then there may be filtering connected on the output - I would remove anything that connects to earth in any way so that it can never go faulty & give you a nasty surprise.
When you do a physical examination, if you think that an extra bit of insulation may be needed consider fitting it.
One isolation transformer I worked on had an output lead rubbing on the case so I fitted a cable tie mount & cable tie so that it was off the case.

I like to test my transformers for insulation before I use them if they haven't been used for a while.

I once talked to a linesman that had worked on High Tension systems where the system was live & he was connected to the conductor.
It was interesting to hear him say that any earth was his enemy & he considered earth like we consider a high voltage potential.


*WARNING: **Some isolation transformers are designed for mains filtering purposes & they DO have the earth connected through.*

These are often made for computers, audio systems, & sensitive electronic equipment.
These use the transformer winding to stop induced transients etc & the couple that I have seen just had the earth connected straight through apart from being connected to some filtering.*
*

*
*

*WARNING:**  Some isolation transformers have been worked on by people who do not know what they are doing & they may connect the earth through.*

I have worked on isolation transformers that suffered from both of those issues that I found on tech benches.
One that I bought was a very nicely made unit that some 'lecky had soldered an extra wire on just to earth the output socket - he did a nice neat job of making the item a death trap.

Some isolation transformers totally isolate the transformer internally & there may not be an earth except on any metal casings but many earth the transformer core for safety but the outlet socket should not have any form of earth wire going to it.
There may even be an internal sign stating to not earth the output.
The couple of isolation transformers I worked on I added a note inside if there wasn't one.


*WARNING:  Only ever connect ONE item to an isolation transformer.*

Older regulations in Australia actually had the output socket surrounded in a shield to stop anyone fitting a "piggyback plug" or double adapter but the modern use of multiple socket boards makes that a bit obsolete.

Isolation transformers for testing also use to have a warning label stating to only connect a single item to the transformer output.


The reason for only connecting a single item is in case of the rare problem of two faulty items being connected to the output.
If one item has an Active (Hot) to metal casing (earth) fault & the other has a Neutral to metal casing (earth) fault then the two items will have full mains potential across their casings.
Because there is no earth connection to trip an protective devices, touching either casing would be safe but touch both at the same time...


Regards,
Brian.

On 14-Mar.-19 00:17, Nicholas Keller wrote:
Raymond,

Per your advice (and that clearly marked on the PCB!), I have purchased an
isolation transformer. After watching a few videos online, I am still
unsure whether I should connect ground from the Tek to the transformer.
One video showed use of a 3-prong to 2-prong adapter where the ground wire
is left unattached. Is this what you describe as “floating the 2247A”?

Thanks for your help!

Nick



On Sun, Mar 3, 2019 at 8:48 PM Nicholas Keller via Groups.Io <Nirokeforums=
gmail.com@groups.io> wrote:

No, I was not using an isolation transformer (I don’t have one) and yes I
was measuring between cap and chassis. I will look into getting a
transformer before testing across c2202 or doing other troubleshooting.

Thanks for the warning

Nick


On Sun, Mar 3, 2019 at 6:49 PM Raymond Domp Frank <@Raymond>
wrote:

On Mon, Mar 4, 2019 at 12:17 AM, Nicholas Keller wrote:

I measured 120-122VAC up to the diode ring. Does this
suggest that C2202 is bad?
I guess you realise you're measuring in an area that is directly
connected
to the mains, no isolation? Take great care! *Do* use an isolation
transformer and *float your 2247A*.

Did you measure the 77 V across C2202 or between C2202 and chassis? The
latter would be incorrect, because the chassis is floating (with
exception
of some C's) against C2202 (since that has a galvanic coupling to the
mains).

Raymond




 

On Thu, Mar 14, 2019 at 05:18 AM, Brian Symons wrote:


*Re: Isolation Transformers*

When using an isolation transformer for troubleshooting you do not want an
earth being connected through.
The whole idea is to break any reference between earth & the mains.
The winding will make both outputs fully isolated from earth so that you can
measure them safely.
This is one time that earth is your enemy.
Brian,
Let's not confuse the OP.
He has a 'scope with a HV (mains switching power supply) part that is galvanically connected to the mains and a LV part that is under normal conditions separated from the HV part through an internal "HF" transformer (inverter output).
Safety earth is normally connected to the 'scope's chassis, not the HV part.

The OP wants to do tests in the rather complicated HV part.
By earthing the chassis and the LV part, the HV part stays isolated.

For all intents and purposes, the LV part is a second isolated part (internal transformer) and there's no reason to float the chassis.

Raymond

Brian Symons
 

I was only answering to the comment about isolation transformers.

If any user gets one then I hope they can check them out to be sure that they are safe instead of expecting it to work & getting bitten.

Having purchased one that was dangerous - a 'lecky had obviously "fixed it to make it safe" which could have killed a user & finding two on test benches that were also dangerous, I just wished to make a purchaser or user aware of possible dangers that may not have been apparent.

I used to do quite a bit of "test & tag" & I found a hell of a lot of dangerous gear even on techs benches.
One of the jobs had several isolation transformers with the through earth of various sizes used for mains filtration so that's how I learned about them.
I saw one of them listed a while ago on eBay as a "test equipment" which it obviously isn't & other users may not have been aware of isolation transformers that aren't designed for the service industry..

Regards,
Brian.

On 14-Mar.-19 22:57, Raymond Domp Frank wrote:
On Thu, Mar 14, 2019 at 05:18 AM, Brian Symons wrote:

*Re: Isolation Transformers*

When using an isolation transformer for troubleshooting you do not want an
earth being connected through.
The whole idea is to break any reference between earth & the mains.
The winding will make both outputs fully isolated from earth so that you can
measure them safely.
This is one time that earth is your enemy.
Brian,
Let's not confuse the OP.
He has a 'scope with a HV (mains switching power supply) part that is galvanically connected to the mains and a LV part that is under normal conditions separated from the HV part through an internal "HF" transformer (inverter output).
Safety earth is normally connected to the 'scope's chassis, not the HV part.

The OP wants to do tests in the rather complicated HV part.
By earthing the chassis and the LV part, the HV part stays isolated.

For all intents and purposes, the LV part is a second isolated part (internal transformer) and there's no reason to float the chassis.

Raymond


John Williams
 

Interesting information Brian. I have often wondered if a Variac would provide isolation also. I have never checked mine I probably should. Regards John

 

On Thu, Mar 14, 2019 at 05:22 PM, John Williams wrote:


I have often wondered if a Variac would provide isolation also.
In general, no.
A Variac is just a primary with an adjustable tap.
Variacs with isolation exist, having their adjustable tap on a secondary winding.

Raymond