Topics

TDS510a Power Supply


Derek Chauran
 

Hello - I am new to this group and generally am very much a beginner with electronics. I recently acquired a couple of TDS scopes - a 640a and a 510a.

The 510a has a dead power supply. I went through the troubleshooting steps in the manual and I have no voltage on the output pins at all. Visual inspection revealed some nasty burn marks around 2 TVS diodes. I don't have a decent way to test them, as I understand with bidirectional TVS diodes, you need a way to apply vrm/vbr to them and I don't have a good way to do this (they're 220 and 250v respectively), so I ordered replacements (they're cheap), but want to make sure I am not just going to fry them again.

There are a couple of dark spots on the board near other components as well. One is a set of resistors (they test correct) parallel with a rectifier diode (I think I have to remove this to test it, as the resistors mess with the diode test) and next to a film capacitor. The other spot is under a bunch of passives and diodes. The resistors and the diodes I can test in circuit seem good as well. Caps look and smell fine, I'd rather not re-cap until I can either fix the supply or confirm that caps are the problem.

I cannot seem to find a schematic for this power supply anywhere. As I am a beginner, and this is a fairly complex power supply, I'm not super sure where to begin without a schematic. I don't want to pull any more components until I have a good test procedure in mind, and perhaps an idea as to what fried it. They were nice enough to mark which side of the board is the primary, and the issues are all on the primary side. Any ideas what to check and where? Any pointers to a schematic, even a partial cobbled together one? I think the 640a has the same power supply, so if I must I can use it to reference voltages at various places.

Pictures of the power supply and scorched areas here: [https://outlook-2.cdn.office.net/assets/mail/file-icon/png/folder_16x16.png] TDS510a<https://1drv.ms/u/s!Ah-c5u-bU1PAq54jn-RzJ6iqx_IDQw?e=M3wyoj> Sorry for the dust, I haven't had a chance to take it outside for a proper cleaning.

One other request - I thought the 2 TVS diodes were identical when I pulled them, but realized that one is a 5k220CA and the other is a 5k250CA. I thought I took a picture before I pulled them, but apparently not. Does anyone have this supply out and can tell me which goes in which position?

Thanks,
Derek


Siggi
 

Hey Derek,

I went through a similar process with the PSU from my TDS684A.
There are schematics to be had that are quite near, if not precisely
correct.
Håkan has this scan on his site:
http://www.hakanh.com/dl/docs/hardtofind/TDS544A_A17-A19(PS).pdf.
Artek manuals has the full service manual for the TDS540, and Tek has the
TDS520B component service manual.

The TVS diodes are snubbers for the standby supply, and they die because a
capacitor (C17?) goes high ESR, which drives up the current through the
snubber until it cooks. There is a relatively recent thread on this here or
on EEVBlog. I simulated the circuit to figure this out on a hint from
another member, as I couldn’t reason this out in my head :).
The two TVS diodes are in series, so it doesn’t matter which way you
install them.
These diodes can go open, or short, or short in one direction (make like a
zener with a diode drop in one direction). If they go open, the switching
transistor will almost certainly die of flyback spike overvoltage. In my
PSU it was a BU508A. Find it and test it, or just replace it.

NB: be careful in there, this supply has a power factor correction switcher
up front, and it runs about 400V on the bulk cap. Good news is that the
supply starts on about 85-90V DC, so if you have a lab supply that goes to
there, you can run it isolated with a current limited source. I seem to
remember there’s an undervoltage protection you can disable in the PFC to
start it even lower under no load conditions.
These supplies do not regulate under no load in my experience, but you can
at least bring up the standby supply.

Siggi

Þann þri., 3. nóv. 2020 kl. 18:23 skrifaði Derek Chauran <af7ux@outlook.com

:
Hello - I am new to this group and generally am very much a beginner with
electronics. I recently acquired a couple of TDS scopes - a 640a and a 510a.

The 510a has a dead power supply. I went through the troubleshooting steps
in the manual and I have no voltage on the output pins at all. Visual
inspection revealed some nasty burn marks around 2 TVS diodes. I don't have
a decent way to test them, as I understand with bidirectional TVS diodes,
you need a way to apply vrm/vbr to them and I don't have a good way to do
this (they're 220 and 250v respectively), so I ordered replacements
(they're cheap), but want to make sure I am not just going to fry them
again.

There are a couple of dark spots on the board near other components as
well. One is a set of resistors (they test correct) parallel with a
rectifier diode (I think I have to remove this to test it, as the resistors
mess with the diode test) and next to a film capacitor. The other spot is
under a bunch of passives and diodes. The resistors and the diodes I can
test in circuit seem good as well. Caps look and smell fine, I'd rather not
re-cap until I can either fix the supply or confirm that caps are the
problem.

I cannot seem to find a schematic for this power supply anywhere. As I am
a beginner, and this is a fairly complex power supply, I'm not super sure
where to begin without a schematic. I don't want to pull any more
components until I have a good test procedure in mind, and perhaps an idea
as to what fried it. They were nice enough to mark which side of the board
is the primary, and the issues are all on the primary side. Any ideas what
to check and where? Any pointers to a schematic, even a partial cobbled
together one? I think the 640a has the same power supply, so if I must I
can use it to reference voltages at various places.

Pictures of the power supply and scorched areas here: [
https://outlook-2.cdn.office.net/assets/mail/file-icon/png/folder_16x16.png]
TDS510a<https://1drv.ms/u/s!Ah-c5u-bU1PAq54jn-RzJ6iqx_IDQw?e=M3wyoj>
Sorry for the dust, I haven't had a chance to take it outside for a proper
cleaning.

One other request - I thought the 2 TVS diodes were identical when I
pulled them, but realized that one is a 5k220CA and the other is a 5k250CA.
I thought I took a picture before I pulled them, but apparently not. Does
anyone have this supply out and can tell me which goes in which position?

Thanks,
Derek






Derek Chauran
 

Thank you Siggy! That 544 schematic does look pretty close. There seem to be a few differences, but on the primary side it's close enough to help me in the right direction. Interestingly, the 520b is in some ways a closer match, but I think it lacks this portion of the supply.

C17 does indeed look to be a good candidate, and is located where one of the darkened patches is. I have an esr meter, so I'd like to verify that it looks bad before I go trying to turn it on again. If I can get it running, and that cap was the problem I'll recap everything.

Point taken about the 400v. I'm using a cat iii meter to measure and am careful not to touch anything. Unfortunately I don't have a bench supply to test with.

Anyhow, will report back when I have some more results.

Thanks,
Derek



________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Siggi <siggi@undo.com>
Sent: Tuesday, November 3, 2020 4:49:30 PM
To: TekScopes@groups.io <TekScopes@groups.io>
Subject: Re: [TekScopes] TDS510a Power Supply

Hey Derek,

I went through a similar process with the PSU from my TDS684A.
There are schematics to be had that are quite near, if not precisely
correct.
Håkan has this scan on his site:
https://eur04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.hakanh.com%2Fdl%2Fdocs%2Fhardtofind%2FTDS544A_A17-A19&;amp;data=04%7C01%7C%7Cde63777228f44ab00caa08d8805b89f4%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400477926093876%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=3uVIwddV2y8fmx2l6w3sw54wDD6%2B7k6dPNAQjFDL%2BjA%3D&amp;reserved=0(PS).pdf.
Artek manuals has the full service manual for the TDS540, and Tek has the
TDS520B component service manual.

The TVS diodes are snubbers for the standby supply, and they die because a
capacitor (C17?) goes high ESR, which drives up the current through the
snubber until it cooks. There is a relatively recent thread on this here or
on EEVBlog. I simulated the circuit to figure this out on a hint from
another member, as I couldn’t reason this out in my head :).
The two TVS diodes are in series, so it doesn’t matter which way you
install them.
These diodes can go open, or short, or short in one direction (make like a
zener with a diode drop in one direction). If they go open, the switching
transistor will almost certainly die of flyback spike overvoltage. In my
PSU it was a BU508A. Find it and test it, or just replace it.

NB: be careful in there, this supply has a power factor correction switcher
up front, and it runs about 400V on the bulk cap. Good news is that the
supply starts on about 85-90V DC, so if you have a lab supply that goes to
there, you can run it isolated with a current limited source. I seem to
remember there’s an undervoltage protection you can disable in the PFC to
start it even lower under no load conditions.
These supplies do not regulate under no load in my experience, but you can
at least bring up the standby supply.

Siggi

Þann þri., 3. nóv. 2020 kl. 18:23 skrifaði Derek Chauran <af7ux@outlook.com
:
Hello - I am new to this group and generally am very much a beginner with
electronics. I recently acquired a couple of TDS scopes - a 640a and a 510a.

The 510a has a dead power supply. I went through the troubleshooting steps
in the manual and I have no voltage on the output pins at all. Visual
inspection revealed some nasty burn marks around 2 TVS diodes. I don't have
a decent way to test them, as I understand with bidirectional TVS diodes,
you need a way to apply vrm/vbr to them and I don't have a good way to do
this (they're 220 and 250v respectively), so I ordered replacements
(they're cheap), but want to make sure I am not just going to fry them
again.

There are a couple of dark spots on the board near other components as
well. One is a set of resistors (they test correct) parallel with a
rectifier diode (I think I have to remove this to test it, as the resistors
mess with the diode test) and next to a film capacitor. The other spot is
under a bunch of passives and diodes. The resistors and the diodes I can
test in circuit seem good as well. Caps look and smell fine, I'd rather not
re-cap until I can either fix the supply or confirm that caps are the
problem.

I cannot seem to find a schematic for this power supply anywhere. As I am
a beginner, and this is a fairly complex power supply, I'm not super sure
where to begin without a schematic. I don't want to pull any more
components until I have a good test procedure in mind, and perhaps an idea
as to what fried it. They were nice enough to mark which side of the board
is the primary, and the issues are all on the primary side. Any ideas what
to check and where? Any pointers to a schematic, even a partial cobbled
together one? I think the 640a has the same power supply, so if I must I
can use it to reference voltages at various places.

Pictures of the power supply and scorched areas here: [
https://eur04.safelinks.protection.outlook.com/?url=https%3A%2F%2Foutlook-2.cdn.office.net%2Fassets%2Fmail%2Ffile-icon%2Fpng%2Ffolder_16x16.png&;amp;data=04%7C01%7C%7Cde63777228f44ab00caa08d8805b89f4%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400477926093876%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=w2TCfVO4c8Rr4Iv1Osu1F9Rf7y%2F7MPc5Pua%2BCfHT5Zo%3D&amp;reserved=0]
TDS510a<https://eur04.safelinks.protection.outlook.com/?url=https%3A%2F%2F1drv.ms%2Fu%2Fs!Ah-c5u-bU1PAq54jn-RzJ6iqx_IDQw%3Fe%3DM3wyoj&amp;data=04%7C01%7C%7Cde63777228f44ab00caa08d8805b89f4%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400477926103871%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=YAWscy96TQBlMSxhUdrZeNObh7LAoAwY0qTfrr39Qpo%3D&amp;reserved=0>
Sorry for the dust, I haven't had a chance to take it outside for a proper
cleaning.

One other request - I thought the 2 TVS diodes were identical when I
pulled them, but realized that one is a 5k220CA and the other is a 5k250CA.
I thought I took a picture before I pulled them, but apparently not. Does
anyone have this supply out and can tell me which goes in which position?

Thanks,
Derek






Siggi
 

Hey Derek,

Here’s the thread I mentioned:
https://groups.io/g/TekScopes/message/160073.

Siggi

Þann mið., 4. nóv. 2020 kl. 05:41 skrifaði Derek Chauran <af7ux@outlook.com

:
Thank you Siggy! That 544 schematic does look pretty close. There seem to
be a few differences, but on the primary side it's close enough to help me
in the right direction. Interestingly, the 520b is in some ways a closer
match, but I think it lacks this portion of the supply.

C17 does indeed look to be a good candidate, and is located where one of
the darkened patches is. I have an esr meter, so I'd like to verify that
it looks bad before I go trying to turn it on again. If I can get it
running, and that cap was the problem I'll recap everything.

Point taken about the 400v. I'm using a cat iii meter to measure and am
careful not to touch anything. Unfortunately I don't have a bench supply to
test with.

Anyhow, will report back when I have some more results.

Thanks,
Derek



________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Siggi <
siggi@undo.com>
Sent: Tuesday, November 3, 2020 4:49:30 PM
To: TekScopes@groups.io <TekScopes@groups.io>
Subject: Re: [TekScopes] TDS510a Power Supply

Hey Derek,

I went through a similar process with the PSU from my TDS684A.
There are schematics to be had that are quite near, if not precisely
correct.
Håkan has this scan on his site:

https://eur04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.hakanh.com%2Fdl%2Fdocs%2Fhardtofind%2FTDS544A_A17-A19&;amp;data=04%7C01%7C%7Cde63777228f44ab00caa08d8805b89f4%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400477926093876%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=3uVIwddV2y8fmx2l6w3sw54wDD6%2B7k6dPNAQjFDL%2BjA%3D&amp;reserved=0(PS).pdf
.
Artek manuals has the full service manual for the TDS540, and Tek has the
TDS520B component service manual.

The TVS diodes are snubbers for the standby supply, and they die because a
capacitor (C17?) goes high ESR, which drives up the current through the
snubber until it cooks. There is a relatively recent thread on this here or
on EEVBlog. I simulated the circuit to figure this out on a hint from
another member, as I couldn’t reason this out in my head :).
The two TVS diodes are in series, so it doesn’t matter which way you
install them.
These diodes can go open, or short, or short in one direction (make like a
zener with a diode drop in one direction). If they go open, the switching
transistor will almost certainly die of flyback spike overvoltage. In my
PSU it was a BU508A. Find it and test it, or just replace it.

NB: be careful in there, this supply has a power factor correction switcher
up front, and it runs about 400V on the bulk cap. Good news is that the
supply starts on about 85-90V DC, so if you have a lab supply that goes to
there, you can run it isolated with a current limited source. I seem to
remember there’s an undervoltage protection you can disable in the PFC to
start it even lower under no load conditions.
These supplies do not regulate under no load in my experience, but you can
at least bring up the standby supply.

Siggi

Þann þri., 3. nóv. 2020 kl. 18:23 skrifaði Derek Chauran <
af7ux@outlook.com
:
Hello - I am new to this group and generally am very much a beginner with
electronics. I recently acquired a couple of TDS scopes - a 640a and a
510a.

The 510a has a dead power supply. I went through the troubleshooting
steps
in the manual and I have no voltage on the output pins at all. Visual
inspection revealed some nasty burn marks around 2 TVS diodes. I don't
have
a decent way to test them, as I understand with bidirectional TVS diodes,
you need a way to apply vrm/vbr to them and I don't have a good way to do
this (they're 220 and 250v respectively), so I ordered replacements
(they're cheap), but want to make sure I am not just going to fry them
again.

There are a couple of dark spots on the board near other components as
well. One is a set of resistors (they test correct) parallel with a
rectifier diode (I think I have to remove this to test it, as the
resistors
mess with the diode test) and next to a film capacitor. The other spot is
under a bunch of passives and diodes. The resistors and the diodes I can
test in circuit seem good as well. Caps look and smell fine, I'd rather
not
re-cap until I can either fix the supply or confirm that caps are the
problem.

I cannot seem to find a schematic for this power supply anywhere. As I am
a beginner, and this is a fairly complex power supply, I'm not super sure
where to begin without a schematic. I don't want to pull any more
components until I have a good test procedure in mind, and perhaps an
idea
as to what fried it. They were nice enough to mark which side of the
board
is the primary, and the issues are all on the primary side. Any ideas
what
to check and where? Any pointers to a schematic, even a partial cobbled
together one? I think the 640a has the same power supply, so if I must I
can use it to reference voltages at various places.

Pictures of the power supply and scorched areas here: [
https://eur04.safelinks.protection.outlook.com/?url=https%3A%2F%2Foutlook-2.cdn.office.net%2Fassets%2Fmail%2Ffile-icon%2Fpng%2Ffolder_16x16.png&;amp;data=04%7C01%7C%7Cde63777228f44ab00caa08d8805b89f4%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400477926093876%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=w2TCfVO4c8Rr4Iv1Osu1F9Rf7y%2F7MPc5Pua%2BCfHT5Zo%3D&amp;reserved=0
]
TDS510a<
https://eur04.safelinks.protection.outlook.com/?url=https%3A%2F%2F1drv.ms%2Fu%2Fs!Ah-c5u-bU1PAq54jn-RzJ6iqx_IDQw%3Fe%3DM3wyoj&amp;data=04%7C01%7C%7Cde63777228f44ab00caa08d8805b89f4%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400477926103871%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=YAWscy96TQBlMSxhUdrZeNObh7LAoAwY0qTfrr39Qpo%3D&amp;reserved=0

Sorry for the dust, I haven't had a chance to take it outside for a
proper
cleaning.

One other request - I thought the 2 TVS diodes were identical when I
pulled them, but realized that one is a 5k220CA and the other is a
5k250CA.
I thought I took a picture before I pulled them, but apparently not. Does
anyone have this supply out and can tell me which goes in which position?

Thanks,
Derek















Siggi
 

Þann mið., 4. nóv. 2020 kl. 05:41 skrifaði Derek Chauran <af7ux@outlook.com
:
Thank you Siggy! That 544 schematic does look pretty close. There seem to
be a few differences, but on the primary side it's close enough to help me
in the right direction. Interestingly, the 520b is in some ways a closer
match, but I think it lacks this portion of the supply.

C17 does indeed look to be a good candidate, and is located where one of
the darkened patches is. I have an esr meter, so I'd like to verify that
it looks bad before I go trying to turn it on again. If I can get it
running, and that cap was the problem I'll recap everything.

Point taken about the 400v. I'm using a cat iii meter to measure and am
careful not to touch anything. Unfortunately I don't have a bench supply to
test with.

If you have any kind of lab supply, the way to defeat the PFC undervoltage
is by feeding a voltage to C1 on the PFC control board, or shunting R15/R20
until the voltage on C1 is high enough.
It’s just so much safer to work that way, and you can probe around the
primary side with a ground referenced instrument like a scope.

BTW: I remember putting a 10X probe on the collector of the switching
transistor. That was unwise, as the swing there is near 1000V when the TVS
are whole! Presumably much higher when they are open.



Anyhow, will report back when I have some more results.

Thanks,
Derek



________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Siggi <
siggi@undo.com>
Sent: Tuesday, November 3, 2020 4:49:30 PM
To: TekScopes@groups.io <TekScopes@groups.io>
Subject: Re: [TekScopes] TDS510a Power Supply

Hey Derek,

I went through a similar process with the PSU from my TDS684A.
There are schematics to be had that are quite near, if not precisely
correct.
Håkan has this scan on his site:

https://eur04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.hakanh.com%2Fdl%2Fdocs%2Fhardtofind%2FTDS544A_A17-A19&;amp;data=04%7C01%7C%7Cde63777228f44ab00caa08d8805b89f4%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400477926093876%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=3uVIwddV2y8fmx2l6w3sw54wDD6%2B7k6dPNAQjFDL%2BjA%3D&amp;reserved=0(PS).pdf
.
Artek manuals has the full service manual for the TDS540, and Tek has the
TDS520B component service manual.

The TVS diodes are snubbers for the standby supply, and they die because a
capacitor (C17?) goes high ESR, which drives up the current through the
snubber until it cooks. There is a relatively recent thread on this here or
on EEVBlog. I simulated the circuit to figure this out on a hint from
another member, as I couldn’t reason this out in my head :).
The two TVS diodes are in series, so it doesn’t matter which way you
install them.
These diodes can go open, or short, or short in one direction (make like a
zener with a diode drop in one direction). If they go open, the switching
transistor will almost certainly die of flyback spike overvoltage. In my
PSU it was a BU508A. Find it and test it, or just replace it.

NB: be careful in there, this supply has a power factor correction switcher
up front, and it runs about 400V on the bulk cap. Good news is that the
supply starts on about 85-90V DC, so if you have a lab supply that goes to
there, you can run it isolated with a current limited source. I seem to
remember there’s an undervoltage protection you can disable in the PFC to
start it even lower under no load conditions.
These supplies do not regulate under no load in my experience, but you can
at least bring up the standby supply.

Siggi

Þann þri., 3. nóv. 2020 kl. 18:23 skrifaði Derek Chauran <
af7ux@outlook.com
:
Hello - I am new to this group and generally am very much a beginner with
electronics. I recently acquired a couple of TDS scopes - a 640a and a
510a.

The 510a has a dead power supply. I went through the troubleshooting
steps
in the manual and I have no voltage on the output pins at all. Visual
inspection revealed some nasty burn marks around 2 TVS diodes. I don't
have
a decent way to test them, as I understand with bidirectional TVS diodes,
you need a way to apply vrm/vbr to them and I don't have a good way to do
this (they're 220 and 250v respectively), so I ordered replacements
(they're cheap), but want to make sure I am not just going to fry them
again.

There are a couple of dark spots on the board near other components as
well. One is a set of resistors (they test correct) parallel with a
rectifier diode (I think I have to remove this to test it, as the
resistors
mess with the diode test) and next to a film capacitor. The other spot is
under a bunch of passives and diodes. The resistors and the diodes I can
test in circuit seem good as well. Caps look and smell fine, I'd rather
not
re-cap until I can either fix the supply or confirm that caps are the
problem.

I cannot seem to find a schematic for this power supply anywhere. As I am
a beginner, and this is a fairly complex power supply, I'm not super sure
where to begin without a schematic. I don't want to pull any more
components until I have a good test procedure in mind, and perhaps an
idea
as to what fried it. They were nice enough to mark which side of the
board
is the primary, and the issues are all on the primary side. Any ideas
what
to check and where? Any pointers to a schematic, even a partial cobbled
together one? I think the 640a has the same power supply, so if I must I
can use it to reference voltages at various places.

Pictures of the power supply and scorched areas here: [
https://eur04.safelinks.protection.outlook.com/?url=https%3A%2F%2Foutlook-2.cdn.office.net%2Fassets%2Fmail%2Ffile-icon%2Fpng%2Ffolder_16x16.png&;amp;data=04%7C01%7C%7Cde63777228f44ab00caa08d8805b89f4%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400477926093876%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=w2TCfVO4c8Rr4Iv1Osu1F9Rf7y%2F7MPc5Pua%2BCfHT5Zo%3D&amp;reserved=0
]
TDS510a<
https://eur04.safelinks.protection.outlook.com/?url=https%3A%2F%2F1drv.ms%2Fu%2Fs!Ah-c5u-bU1PAq54jn-RzJ6iqx_IDQw%3Fe%3DM3wyoj&amp;data=04%7C01%7C%7Cde63777228f44ab00caa08d8805b89f4%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400477926103871%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=YAWscy96TQBlMSxhUdrZeNObh7LAoAwY0qTfrr39Qpo%3D&amp;reserved=0

Sorry for the dust, I haven't had a chance to take it outside for a
proper
cleaning.

One other request - I thought the 2 TVS diodes were identical when I
pulled them, but realized that one is a 5k220CA and the other is a
5k250CA.
I thought I took a picture before I pulled them, but apparently not. Does
anyone have this supply out and can tell me which goes in which position?

Thanks,
Derek















Derek Chauran
 

Oh, my, well, I think 7.2 ohm is a bit high for the ESR on that cap. The one next to it is around 1.6 too. i was going to re-cap the thing if I got it working, but now am thinking I should just replace all of the caps now. The big 450v cap looks maybe a tiny bit bulged.

I also have a bu508a and my little tester says it's a resistor between the collector and emitter, rather than a transistor, so that's bad as well. Thankfully that portion of the supply is pretty simple , and I can rebuild it for a few $, and recap it for not too many more.



________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Derek Chauran <af7ux@outlook.com>
Sent: Wednesday, November 4, 2020 2:41 AM
To: tekscopes@groups.io <tekscopes@groups.io>
Subject: Re: [TekScopes] TDS510a Power Supply

Thank you Siggy! That 544 schematic does look pretty close. There seem to be a few differences, but on the primary side it's close enough to help me in the right direction. Interestingly, the 520b is in some ways a closer match, but I think it lacks this portion of the supply.

C17 does indeed look to be a good candidate, and is located where one of the darkened patches is. I have an esr meter, so I'd like to verify that it looks bad before I go trying to turn it on again. If I can get it running, and that cap was the problem I'll recap everything.

Point taken about the 400v. I'm using a cat iii meter to measure and am careful not to touch anything. Unfortunately I don't have a bench supply to test with.

Anyhow, will report back when I have some more results.

Thanks,
Derek



________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Siggi <siggi@undo.com>
Sent: Tuesday, November 3, 2020 4:49:30 PM
To: TekScopes@groups.io <TekScopes@groups.io>
Subject: Re: [TekScopes] TDS510a Power Supply

Hey Derek,

I went through a similar process with the PSU from my TDS684A.
There are schematics to be had that are quite near, if not precisely
correct.
Håkan has this scan on his site:
https://eur05.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.hakanh.com%2Fdl%2Fdocs%2Fhardtofind%2FTDS544A_A17-A19&;amp;data=04%7C01%7C%7C3bf19d640d08493dc23608d880ae3500%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400832981847223%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=dYIiMMiez0slwqWdPANtDUKKicvYzFtGolYkxsZosaw%3D&amp;reserved=0(PS).pdf.
Artek manuals has the full service manual for the TDS540, and Tek has the
TDS520B component service manual.

The TVS diodes are snubbers for the standby supply, and they die because a
capacitor (C17?) goes high ESR, which drives up the current through the
snubber until it cooks. There is a relatively recent thread on this here or
on EEVBlog. I simulated the circuit to figure this out on a hint from
another member, as I couldn’t reason this out in my head :).
The two TVS diodes are in series, so it doesn’t matter which way you
install them.
These diodes can go open, or short, or short in one direction (make like a
zener with a diode drop in one direction). If they go open, the switching
transistor will almost certainly die of flyback spike overvoltage. In my
PSU it was a BU508A. Find it and test it, or just replace it.

NB: be careful in there, this supply has a power factor correction switcher
up front, and it runs about 400V on the bulk cap. Good news is that the
supply starts on about 85-90V DC, so if you have a lab supply that goes to
there, you can run it isolated with a current limited source. I seem to
remember there’s an undervoltage protection you can disable in the PFC to
start it even lower under no load conditions.
These supplies do not regulate under no load in my experience, but you can
at least bring up the standby supply.

Siggi

Þann þri., 3. nóv. 2020 kl. 18:23 skrifaði Derek Chauran <af7ux@outlook.com
:
Hello - I am new to this group and generally am very much a beginner with
electronics. I recently acquired a couple of TDS scopes - a 640a and a 510a.

The 510a has a dead power supply. I went through the troubleshooting steps
in the manual and I have no voltage on the output pins at all. Visual
inspection revealed some nasty burn marks around 2 TVS diodes. I don't have
a decent way to test them, as I understand with bidirectional TVS diodes,
you need a way to apply vrm/vbr to them and I don't have a good way to do
this (they're 220 and 250v respectively), so I ordered replacements
(they're cheap), but want to make sure I am not just going to fry them
again.

There are a couple of dark spots on the board near other components as
well. One is a set of resistors (they test correct) parallel with a
rectifier diode (I think I have to remove this to test it, as the resistors
mess with the diode test) and next to a film capacitor. The other spot is
under a bunch of passives and diodes. The resistors and the diodes I can
test in circuit seem good as well. Caps look and smell fine, I'd rather not
re-cap until I can either fix the supply or confirm that caps are the
problem.

I cannot seem to find a schematic for this power supply anywhere. As I am
a beginner, and this is a fairly complex power supply, I'm not super sure
where to begin without a schematic. I don't want to pull any more
components until I have a good test procedure in mind, and perhaps an idea
as to what fried it. They were nice enough to mark which side of the board
is the primary, and the issues are all on the primary side. Any ideas what
to check and where? Any pointers to a schematic, even a partial cobbled
together one? I think the 640a has the same power supply, so if I must I
can use it to reference voltages at various places.

Pictures of the power supply and scorched areas here: [
https://eur05.safelinks.protection.outlook.com/?url=https%3A%2F%2Foutlook-2.cdn.office.net%2Fassets%2Fmail%2Ffile-icon%2Fpng%2Ffolder_16x16.png&;amp;data=04%7C01%7C%7C3bf19d640d08493dc23608d880ae3500%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400832981857218%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=tXmZdAnyQ7PpTyAEwANBgEucrivOpTRXesr5citSOuc%3D&amp;reserved=0]
TDS510a<https://eur05.safelinks.protection.outlook.com/?url=https%3A%2F%2F1drv.ms%2Fu%2Fs!Ah-c5u-bU1PAq54jn-RzJ6iqx_IDQw%3Fe%3DM3wyoj&amp;data=04%7C01%7C%7C3bf19d640d08493dc23608d880ae3500%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637400832981857218%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=NIhG3LRTA%2FtvaY7VQWDXCKNPU3LKSLLg16gLVwgAz6s%3D&amp;reserved=0>
Sorry for the dust, I haven't had a chance to take it outside for a proper
cleaning.

One other request - I thought the 2 TVS diodes were identical when I
pulled them, but realized that one is a 5k220CA and the other is a 5k250CA.
I thought I took a picture before I pulled them, but apparently not. Does
anyone have this supply out and can tell me which goes in which position?

Thanks,
Derek






Siggi
 

On Wed, Nov 4, 2020 at 10:15 PM Derek Chauran <af7ux@outlook.com> wrote:

Oh, my, well, I think 7.2 ohm is a bit high for the ESR on that cap.

Yeah, this cap sees a fair bit of ripple current, so wants to be low ESR.
As this is in the standby supply, unless you turn off the hard switch at
the back, it'll be under stress 24/7.


The one next to it is around 1.6 too. i was going to re-cap the thing if I
got it working, but now am thinking I should just replace all of the caps
now. The big 450v cap looks maybe a tiny bit bulged.
So did mine, but it was perfectly fine. I don't think the DC bus bulk cap
is under a great deal of stress.


I also have a bu508a and my little tester says it's a resistor between the
collector and emitter, rather than a transistor, so that's bad as well.

That'd take down your standby supply for sure :).


Derek Chauran
 

Thanks Siggi! That cap is supposed to be less than 0.3 ohm, and I ordered a low ESR cap to replace it. I have ordered all of the parts to fix this, and I went ahead and ordered the caps to re-cap the whole thing. It costs me $8 to get stuff shipped from mouser, which is a good chunk of the cost for a full re-cap, and you can never have too many caps if I decide not to do it 😄. That said, the caps in the rest of the scope look really good. All of the solder joints are still nice and shiny, so I probably won't do the whole scope just yet.

I am probably going to sell this scope, the 510 seems to sell for more than the 640, even though the 640 looks like the better scope on-paper.

Thanks,
Derek

________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Siggi <siggi@undo.com>
Sent: Thursday, November 5, 2020 5:19 AM
To: TekScopes@groups.io <TekScopes@groups.io>
Subject: Re: [TekScopes] TDS510a Power Supply

On Wed, Nov 4, 2020 at 10:15 PM Derek Chauran <af7ux@outlook.com> wrote:

Oh, my, well, I think 7.2 ohm is a bit high for the ESR on that cap.

Yeah, this cap sees a fair bit of ripple current, so wants to be low ESR.
As this is in the standby supply, unless you turn off the hard switch at
the back, it'll be under stress 24/7.


The one next to it is around 1.6 too. i was going to re-cap the thing if I
got it working, but now am thinking I should just replace all of the caps
now. The big 450v cap looks maybe a tiny bit bulged.
So did mine, but it was perfectly fine. I don't think the DC bus bulk cap
is under a great deal of stress.


I also have a bu508a and my little tester says it's a resistor between the
collector and emitter, rather than a transistor, so that's bad as well.

That'd take down your standby supply for sure :).


Siggi
 

On Thu, Nov 5, 2020 at 2:16 PM Derek Chauran <af7ux@outlook.com> wrote:

I am probably going to sell this scope, the 510 seems to sell for more
than the 640, even though the 640 looks like the better scope on-paper.
There are a couple of specs you should probably compare before you decide
which to keep and which to sell.

The sample buffer size might be a killer - you should compare the two,
especially as the TDS6XX don't have peak detect until the B-series (if
memory serves). Between a short record length and no peak detect, they can
be quite useless for slow captures.

I also don't think the TDS600s have averaging, nor hi-res, nor equivalent
time sampling.


Derek Chauran
 

Thanks, as I mentioned, I am a novice at all of this, so it is very helpful to have pointers as to what to look into. I have been reading up based on your comments. Learning is exactly why I am here, and I have learned a lot.

For most of what I can imagine using the scope for, I will be looking at repetitive signals. E.g. checking power supplies for ripple, or observing simple, modulated RF signals (I'm a ham, which is where my interest in electronics stems from). In other words, I am primarily going to be looking at analog signals.

The 640a manual states that it has averaging, but not peak or high resolution.

I did a quick read on equivalent time sampling - the 640 has a higher sample rate (2g/sec) and interpolation, so the net effect is that both scopes claim an "effective" sampling rate of 100gsamples/sec, but the 640 has a better single shot bandwidth, and 2gs/sec on all 4 channels, whereas the 510a has 500ms/sec on 2 channels or 250ms/sec on 4.

The 510a definitely wins in sample buffer size - 50k vs 2k. That said, even Tek suggests that for most signals 2k is plenty.

Regarding peak, it looks like the 640a has envelope mode which is fine for repetitive signals, whereas peak is better for single shot? But that also means that peak is bandwidth limited on the 510a. Curiously I cannot find the non-interpolated real-time bandwidth of the 510a, the manual just says that it will decide when it forces interpolation or equivalent time sampling. I would assume that it's something like 125 mhz?

High res - also a clear winner in the 510a.

Thanks,
Derek
________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Siggi <siggi@undo.com>
Sent: Thursday, November 5, 2020 11:36 AM
To: TekScopes@groups.io <TekScopes@groups.io>
Subject: Re: [TekScopes] TDS510a Power Supply

On Thu, Nov 5, 2020 at 2:16 PM Derek Chauran <af7ux@outlook.com> wrote:

I am probably going to sell this scope, the 510 seems to sell for more
than the 640, even though the 640 looks like the better scope on-paper.
There are a couple of specs you should probably compare before you decide
which to keep and which to sell.

The sample buffer size might be a killer - you should compare the two,
especially as the TDS6XX don't have peak detect until the B-series (if
memory serves). Between a short record length and no peak detect, they can
be quite useless for slow captures.

I also don't think the TDS600s have averaging, nor hi-res, nor equivalent
time sampling.


Siggi
 

On Thu, Nov 5, 2020 at 5:35 PM Derek Chauran <af7ux@outlook.com> wrote:

Thanks, as I mentioned, I am a novice at all of this, so it is very
helpful to have pointers as to what to look into. I have been reading up
based on your comments. Learning is exactly why I am here, and I have
learned a lot.
The good news is that you have a choice of two cutting-edge scopes as
of early 90s. There were good reasons why each was a Tek product at
the time, and which one is better for your needs will totally depend on you
and your needs. If you're not desperate for the cash, I'd suggest you keep
both scopes for a while, on the same bench, and figure out which one you
prefer.

I'm sure there are others here who're better qualified to speak to these
two scopes than I am, but here's my take.

For most of what I can imagine using the scope for, I will be looking at
repetitive signals. E.g. checking power supplies for ripple, or observing
simple, modulated RF signals (I'm a ham, which is where my interest in
electronics stems from). In other words, I am primarily going to be looking
at analog signals.
Neither scope is going to have any trouble with either of the above, but
depending on the frequency of your RF, the TDS510A may have features such
as high-res that will be helpful.


The 640a manual states that it has averaging, but not peak or high
resolution.
Peak detect (or envelope) is handy when you're looking at a "slow" one-time
signal on a scope with record length that's too short to capture all the
details of the signal. I have a 2430, which has only ~1024 sample record
length, and when I'd use it to e.g. look at a switch mode power supply
startup, I'd use envelope or peak detect to get the envelope of the
startup. Best case you'd want infinite record length so you could capture
forever with full detail. Newer scopes approximate that ideal better and
better.
My TDS784D has a max of 8MS record, which is sometimes marginal. The scope
is also slow as molasses when viewing a record that long, which is also
annoying.


I did a quick read on equivalent time sampling - the 640 has a higher
sample rate (2g/sec) and interpolation, so the net effect is that both
scopes claim an "effective" sampling rate of 100gsamples/sec, but the 640
has a better single shot bandwidth, and 2gs/sec on all 4 channels, whereas
the 510a has 500ms/sec on 2 channels or 250ms/sec on 4.
Yeah, the TDS6XX scopes went all out on realtime sample rates at the cost
of whatever else - record length for one. This was done by capturing the
signal into a CCD storage array at a high rate, then digitizing the signal
out of the CCD at a lower rate. I'm not sure they have equivalent time, as
to do this you need some timing hardware to measure the difference from
sample clock to trigger, and I don't remember seeing that (but I'm often
wrong).
There are tradeoffs to the CCD capture, and the TDS6XX series dials all the
knobs to one side. As a case in point, they're practically blind all the
time - meaning that while they capture a single record in realtime, it
takes a fair amount of time to read the capture out of the CCD, store it in
memory and display it.
You might amuse yourself by hooking the same signal source into each scope,
while using the other scope to look at the trigger output on the back. I
think you'll find that under practically all conditions, the TDS6XX scope
will trigger at a lower rate than the TDS510A.

In the later TDSen this "blindness" - even in the TDS5/7XXen - was deemed
important enough that the DPO feature was implemented. This dials all the
knobs in the direction of minimal blindness to allow capturing and
displaying glitches and other anomalies at the highest possible
probability. I find this occasionally convenient, as by turning on DPO and
infinite persistence, I'm reasonably assured that I'll see glitching
"sooner" than by infinite persistence alone.

The 510a definitely wins in sample buffer size - 50k vs 2k. That said, even
Tek suggests that for most signals 2k is plenty.
Mmmm, yeah, I believe the literature of the time would have said that. I
don't think you'd find many people agreeing with that today. Typically I'd
choose record length over realtime bandwidth - except when I need the
realtime bandwidth :).
On my bench I have a 2467, a HP54622D and a TDS784D, which I use in that
order of preference (mind you, I'm but a hobbyist with a bit of a gear
fetish).
The 2467 is an analog scope that's a total joy to use for repetitive
signals. For slow one-time signals, I prefer the HP, and only when I run
out of bandwidth I turn on the TDS784D to suffer the fans and the UI.

Regarding peak, it looks like the 640a has envelope mode which is fine for
repetitive signals, whereas peak is better for single shot? But that also
means that peak is bandwidth limited on the 510a. Curiously I cannot find
the non-interpolated real-time bandwidth of the 510a, the manual just says
that it will decide when it forces interpolation or equivalent time
sampling.

Yeah, the spec sheets will be shy about their Achilles heels.

I would assume that it's something like 125 mhz?
Ehem, MHz (m is milli :). For sure you don't want to go above 250MHz with
500MS/s.

High res - also a clear winner in the 510a.
I have found this useful on occasion, but again this brings down your
realtime sample rate.


Derek Chauran
 

Hi Siggi, well, the 510a is now working, thanks to your advice! I ended up replacing c16 and c17, the 2 tvs diodes, and the transistor. Recapping the whole thing seemed like more effort than it was worth, and the rest of the caps tested with good values and low esr in circuit, but I have all the caps, just in case.

Thanks for the help! Now on to figuring out how to backup the eeproms
________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Siggi <siggi@undo.com>
Sent: Thursday, November 5, 2020 4:24:16 PM
To: TekScopes@groups.io <TekScopes@groups.io>
Subject: Re: [TekScopes] TDS510a Power Supply

On Thu, Nov 5, 2020 at 5:35 PM Derek Chauran <af7ux@outlook.com> wrote:

Thanks, as I mentioned, I am a novice at all of this, so it is very
helpful to have pointers as to what to look into. I have been reading up
based on your comments. Learning is exactly why I am here, and I have
learned a lot.
The good news is that you have a choice of two cutting-edge scopes as
of early 90s. There were good reasons why each was a Tek product at
the time, and which one is better for your needs will totally depend on you
and your needs. If you're not desperate for the cash, I'd suggest you keep
both scopes for a while, on the same bench, and figure out which one you
prefer.

I'm sure there are others here who're better qualified to speak to these
two scopes than I am, but here's my take.

For most of what I can imagine using the scope for, I will be looking at
repetitive signals. E.g. checking power supplies for ripple, or observing
simple, modulated RF signals (I'm a ham, which is where my interest in
electronics stems from). In other words, I am primarily going to be looking
at analog signals.
Neither scope is going to have any trouble with either of the above, but
depending on the frequency of your RF, the TDS510A may have features such
as high-res that will be helpful.


The 640a manual states that it has averaging, but not peak or high
resolution.
Peak detect (or envelope) is handy when you're looking at a "slow" one-time
signal on a scope with record length that's too short to capture all the
details of the signal. I have a 2430, which has only ~1024 sample record
length, and when I'd use it to e.g. look at a switch mode power supply
startup, I'd use envelope or peak detect to get the envelope of the
startup. Best case you'd want infinite record length so you could capture
forever with full detail. Newer scopes approximate that ideal better and
better.
My TDS784D has a max of 8MS record, which is sometimes marginal. The scope
is also slow as molasses when viewing a record that long, which is also
annoying.


I did a quick read on equivalent time sampling - the 640 has a higher
sample rate (2g/sec) and interpolation, so the net effect is that both
scopes claim an "effective" sampling rate of 100gsamples/sec, but the 640
has a better single shot bandwidth, and 2gs/sec on all 4 channels, whereas
the 510a has 500ms/sec on 2 channels or 250ms/sec on 4.
Yeah, the TDS6XX scopes went all out on realtime sample rates at the cost
of whatever else - record length for one. This was done by capturing the
signal into a CCD storage array at a high rate, then digitizing the signal
out of the CCD at a lower rate. I'm not sure they have equivalent time, as
to do this you need some timing hardware to measure the difference from
sample clock to trigger, and I don't remember seeing that (but I'm often
wrong).
There are tradeoffs to the CCD capture, and the TDS6XX series dials all the
knobs to one side. As a case in point, they're practically blind all the
time - meaning that while they capture a single record in realtime, it
takes a fair amount of time to read the capture out of the CCD, store it in
memory and display it.
You might amuse yourself by hooking the same signal source into each scope,
while using the other scope to look at the trigger output on the back. I
think you'll find that under practically all conditions, the TDS6XX scope
will trigger at a lower rate than the TDS510A.

In the later TDSen this "blindness" - even in the TDS5/7XXen - was deemed
important enough that the DPO feature was implemented. This dials all the
knobs in the direction of minimal blindness to allow capturing and
displaying glitches and other anomalies at the highest possible
probability. I find this occasionally convenient, as by turning on DPO and
infinite persistence, I'm reasonably assured that I'll see glitching
"sooner" than by infinite persistence alone.

The 510a definitely wins in sample buffer size - 50k vs 2k. That said, even
Tek suggests that for most signals 2k is plenty.
Mmmm, yeah, I believe the literature of the time would have said that. I
don't think you'd find many people agreeing with that today. Typically I'd
choose record length over realtime bandwidth - except when I need the
realtime bandwidth :).
On my bench I have a 2467, a HP54622D and a TDS784D, which I use in that
order of preference (mind you, I'm but a hobbyist with a bit of a gear
fetish).
The 2467 is an analog scope that's a total joy to use for repetitive
signals. For slow one-time signals, I prefer the HP, and only when I run
out of bandwidth I turn on the TDS784D to suffer the fans and the UI.

Regarding peak, it looks like the 640a has envelope mode which is fine for
repetitive signals, whereas peak is better for single shot? But that also
means that peak is bandwidth limited on the 510a. Curiously I cannot find
the non-interpolated real-time bandwidth of the 510a, the manual just says
that it will decide when it forces interpolation or equivalent time
sampling.

Yeah, the spec sheets will be shy about their Achilles heels.

I would assume that it's something like 125 mhz?
Ehem, MHz (m is milli :). For sure you don't want to go above 250MHz with
500MS/s.

High res - also a clear winner in the 510a.
I have found this useful on occasion, but again this brings down your
realtime sample rate.


Siggi
 

On Mon, Nov 9, 2020 at 11:01 PM Derek Chauran <af7ux@outlook.com> wrote:

Hi Siggi, well, the 510a is now working, thanks to your advice! I ended up
replacing c16 and c17, the 2 tvs diodes, and the transistor.
Congrats on the fix!