Topics

547 Repair - how do you locate where parts are located on ceramic strips?


Grayson Evans
 

Trying to repair my 547 workhorse. Used it for years. Lost B horiz. display blanking. I know where the problem is but I can't figure out where the heck the parts are located on the ceramic strips. I can get in the neighborhood, but tracing from tube/trans. sockets is damn hard since the parts on the ceramic strips cover up the sockets.
For example, where is R137? I have all the manuals I know of but none have a parts locator diagram. There must be one someplace, right?
Thanks, Grayson, KJ7UM


Albert Otten
 

For example, where is R137?
Thanks, Grayson, KJ7UM
That's an easy one (once you found it!).
Swing panel, middle pair of ceramic strips, between 2nd notches counted from the rear. Value rings brown-black-black-yellow.
As far as I know the only way to find components is what you said. Be aware of the sometimes hardly visible wire bridges between notches.
Albert


Morris Odell
 

This problem has been discussed here in the past. A common loss of B sweep intensification (not actually retrace blanking) is caused by failure of either or both Q184 or Q173, both fast pnp germanium transistors type 2N2207. Fortunately the transistors are socketed so there's no soldering needed to replace them. Q 184 can be replaced easily with a modern silicon pnp transistor capable of withstanding 100 volts or so such as a BC 640 or similar. If Q 173 fails it can also be replaced by a modern silicon pnp but C170 (200 pf) may need to be increased to allow it to switch fast enough at the highest sweep speeds.

Hope this helps,

Morris VK3DOC


Dave Wise
 

For Q173 and Q373, the critical attributes are Cob and BVcbo. Q184 same except Cob is not top priority.
My best results were BC556 and KSA992. Both required extra speedup capacitance. MPSA56 was okay but it needed more C. I don't have a BC640, but it probably does too since Cob is 9pF typical vs 6pF max. I would try 330pF. Be sure you know the pinout of whatever you try, they're all different.

As far as I can tell, no modern transistor equals the 2N2207 in this application.

Here's the conclusion from my original post.

-----------------

RECOMMENDED (demonstrated working)

BC556 + 220pF
KSA992 + 220pF
MPSA56 + 470pF

NOT RECOMMENDED (insufficient breakdown voltage)

2N3906
2N5086
2N5087
KSA1015
PN4133

NOT RECOMMENDED (slow even with cap)

2N5401
BC638
KSA708

GOOD BUT HARD TO GET (obsolete)

AF118
2N2207
MPSH55

NOT TESTED (because I don't have)

2N3486
2N3505
2N3799
2N3962
BC640
KSA733 probably no better than KSA992
PN3645
PN4249

OTHER 2N2207 USAGE

The 547 has 2N2207 in six places: Q173/Q184/Q373/Q424/Q554/Q564. I believe that all
of these can be replaced by BC556 or KSA992, subject to enlarging C170/C370.

-----------------

Regards,
Dave Wise
________________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Morris Odell via groups.io <vilgotch1=gmail.com@groups.io>
Sent: Monday, September 14, 2020 6:29 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] 547 Repair - how do you locate where parts are located on ceramic strips?

This problem has been discussed here in the past. A common loss of B sweep intensification (not actually retrace blanking) is caused by failure of either or both Q184 or Q173, both fast pnp germanium transistors type 2N2207. Fortunately the transistors are socketed so there's no soldering needed to replace them. Q 184 can be replaced easily with a modern silicon pnp transistor capable of withstanding 100 volts or so such as a BC 640 or similar. If Q 173 fails it can also be replaced by a modern silicon pnp but C170 (200 pf) may need to be increased to allow it to switch fast enough at the highest sweep speeds.

Hope this helps,

Morris VK3DOC


Bill Perkins
 

There's a slew of what look to be real deal NOS 2N2207s on eBay, here:

https://tinyurl.com/yyjt6vku

Best,
Bill

For Q173 and Q373, the critical attributes are Cob and BVcbo. Q184 same except Cob is not top priority.
My best results were BC556 and KSA992. Both required extra speedup capacitance. MPSA56 was okay but it needed more C. I don't have a BC640, but it probably does too since Cob is 9pF typical vs 6pF max. I would try 330pF. Be sure you know the pinout of whatever you try, they're all different.
As far as I can tell, no modern transistor equals the 2N2207 in this application.
Here's the conclusion from my original post.
-----------------
RECOMMENDED (demonstrated working)
BC556 + 220pF
KSA992 + 220pF
MPSA56 + 470pF
NOT RECOMMENDED (insufficient breakdown voltage)
2N3906
2N5086
2N5087
KSA1015
PN4133
NOT RECOMMENDED (slow even with cap)
2N5401
BC638
KSA708
GOOD BUT HARD TO GET (obsolete)
AF118
2N2207
MPSH55
NOT TESTED (because I don't have)
2N3486
2N3505
2N3799
2N3962
BC640
KSA733 probably no better than KSA992
PN3645
PN4249
OTHER 2N2207 USAGE
The 547 has 2N2207 in six places: Q173/Q184/Q373/Q424/Q554/Q564. I believe that all
of these can be replaced by BC556 or KSA992, subject to enlarging C170/C370.
-----------------
Regards,
Dave Wise


Dave Wise
 

That will appeal to people who just want their scope working and don't get a big kick out of re-engineering it as I do. I think my hobby niche is becoming "retrofit engineering".

Dave Wise
________________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Bill Perkins via groups.io <sales=pearl-hifi.com@groups.io>
Sent: Tuesday, September 15, 2020 11:59 AM
To: TekScopes@groups.io
Subject: Re: [TekScopes] 547 Repair - how do you locate where parts are located on ceramic strips?

There's a slew of what look to be real deal NOS 2N2207s on eBay, here:

https://tinyurl.com/yyjt6vku

Best,
Bill

For Q173 and Q373, the critical attributes are Cob and BVcbo. Q184 same except Cob is not top priority.
My best results were BC556 and KSA992. Both required extra speedup capacitance. MPSA56 was okay but it needed more C. I don't have a BC640, but it probably does too since Cob is 9pF typical vs 6pF max. I would try 330pF. Be sure you know the pinout of whatever you try, they're all different.

As far as I can tell, no modern transistor equals the 2N2207 in this application.

Here's the conclusion from my original post.

-----------------

RECOMMENDED (demonstrated working)

BC556 + 220pF
KSA992 + 220pF
MPSA56 + 470pF

NOT RECOMMENDED (insufficient breakdown voltage)

2N3906
2N5086
2N5087
KSA1015
PN4133

NOT RECOMMENDED (slow even with cap)

2N5401
BC638
KSA708

GOOD BUT HARD TO GET (obsolete)

AF118
2N2207
MPSH55

NOT TESTED (because I don't have)

2N3486
2N3505
2N3799
2N3962
BC640
KSA733 probably no better than KSA992
PN3645
PN4249

OTHER 2N2207 USAGE

The 547 has 2N2207 in six places: Q173/Q184/Q373/Q424/Q554/Q564. I believe that all
of these can be replaced by BC556 or KSA992, subject to enlarging C170/C370.

-----------------

Regards,
Dave Wise


Tom Lee
 

It is perhaps worth noting that the tin whisker problem is a function of age, not of the time powered up, so NOS transistors in that package have the same likelihood of having whiskers as any other of the same vintage.

Tom

Sent from my iThing, so please forgive brevity and typos

On Sep 15, 2020, at 12:12, Dave Wise <david_wise@...> wrote:

That will appeal to people who just want their scope working and don't get a big kick out of re-engineering it as I do. I think my hobby niche is becoming "retrofit engineering".

Dave Wise
________________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Bill Perkins via groups.io <sales=pearl-hifi.com@groups.io>
Sent: Tuesday, September 15, 2020 11:59 AM
To: TekScopes@groups.io
Subject: Re: [TekScopes] 547 Repair - how do you locate where parts are located on ceramic strips?

There's a slew of what look to be real deal NOS 2N2207s on eBay, here:

https://tinyurl.com/yyjt6vku

Best,
Bill

For Q173 and Q373, the critical attributes are Cob and BVcbo. Q184 same except Cob is not top priority.
My best results were BC556 and KSA992. Both required extra speedup capacitance. MPSA56 was okay but it needed more C. I don't have a BC640, but it probably does too since Cob is 9pF typical vs 6pF max. I would try 330pF. Be sure you know the pinout of whatever you try, they're all different.

As far as I can tell, no modern transistor equals the 2N2207 in this application.

Here's the conclusion from my original post.

-----------------

RECOMMENDED (demonstrated working)

BC556 + 220pF
KSA992 + 220pF
MPSA56 + 470pF

NOT RECOMMENDED (insufficient breakdown voltage)

2N3906
2N5086
2N5087
KSA1015
PN4133

NOT RECOMMENDED (slow even with cap)

2N5401
BC638
KSA708

GOOD BUT HARD TO GET (obsolete)

AF118
2N2207
MPSH55

NOT TESTED (because I don't have)

2N3486
2N3505
2N3799
2N3962
BC640
KSA733 probably no better than KSA992
PN3645
PN4249

OTHER 2N2207 USAGE

The 547 has 2N2207 in six places: Q173/Q184/Q373/Q424/Q554/Q564. I believe that all
of these can be replaced by BC556 or KSA992, subject to enlarging C170/C370.

-----------------

Regards,
Dave Wise







Grayson Evans
 

Thanks a bunch Morris. Yes, that helps! I figured it had to be one of those two transistors, but while they are easy to remove/replace physically, I was trying to find where the EBC leads wound up on the ceramic strip so I could take some measurements, but their sockets and V193 are completely covered by the parts on the ceramic strips. I can't even see them. I wasn't having much luck trying to figure out which parts on the strips matched the schematic. That is why I asked if there was some kind of parts placement diagram someplace.


Grayson Evans
 

Thanks a bunch Dave. That is great info. I will definitely do some experimenting with your recommended substitutes.
I have very little transistor stock since I build/restore tube gear almost exclusively, so off to Digikey.
By "I would try 330pF" I am assuming you mean C170.

Thanks again,
Grayson, KJ7UM


Dave Wise
 

Yes.

Sometimes you can clear the short without also damaging the semiconductor element, and the OP should try it. I've done it, by shorting EBC together and applying voltage between them and the can.

Dave Wise
________________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Tom Lee via groups.io <tomlee=ee.stanford.edu@groups.io>
Sent: Tuesday, September 15, 2020 12:20 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] 547 Repair - how do you locate where parts are located on ceramic strips?

It is perhaps worth noting that the tin whisker problem is a function of age, not of the time powered up, so NOS transistors in that package have the same likelihood of having whiskers as any other of the same vintage.

Tom

Sent from my iThing, so please forgive brevity and typos

On Sep 15, 2020, at 12:12, Dave Wise <david_wise@...> wrote:

That will appeal to people who just want their scope working and don't get a big kick out of re-engineering it as I do. I think my hobby niche is becoming "retrofit engineering".

Dave Wise
________________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Bill Perkins via groups.io <sales=pearl-hifi.com@groups.io>
Sent: Tuesday, September 15, 2020 11:59 AM
To: TekScopes@groups.io
Subject: Re: [TekScopes] 547 Repair - how do you locate where parts are located on ceramic strips?

There's a slew of what look to be real deal NOS 2N2207s on eBay, here:

https://tinyurl.com/yyjt6vku

Best,
Bill

For Q173 and Q373, the critical attributes are Cob and BVcbo. Q184 same except Cob is not top priority.
My best results were BC556 and KSA992. Both required extra speedup capacitance. MPSA56 was okay but it needed more C. I don't have a BC640, but it probably does too since Cob is 9pF typical vs 6pF max. I would try 330pF. Be sure you know the pinout of whatever you try, they're all different.

As far as I can tell, no modern transistor equals the 2N2207 in this application.

Here's the conclusion from my original post.

-----------------

RECOMMENDED (demonstrated working)

BC556 + 220pF
KSA992 + 220pF
MPSA56 + 470pF

NOT RECOMMENDED (insufficient breakdown voltage)

2N3906
2N5086
2N5087
KSA1015
PN4133

NOT RECOMMENDED (slow even with cap)

2N5401
BC638
KSA708

GOOD BUT HARD TO GET (obsolete)

AF118
2N2207
MPSH55

NOT TESTED (because I don't have)

2N3486
2N3505
2N3799
2N3962
BC640
KSA733 probably no better than KSA992
PN3645
PN4249

OTHER 2N2207 USAGE

The 547 has 2N2207 in six places: Q173/Q184/Q373/Q424/Q554/Q564. I believe that all
of these can be replaced by BC556 or KSA992, subject to enlarging C170/C370.

-----------------

Regards,
Dave Wise







Grayson Evans
 

Hello Dave,

I order several BC566's to replace the 2N2207. Do you have any idea what the pin-out is on the 2N2207. I have searched the internet but can't find a data sheet on the 2207. i assume on of the leads is the case, I can't even see the back of the sockets due their being right under the parts on the ceramic strips.
Thanks !

Grayson


Chuck Harris
 

Tektronix usually puts the transistor pin-outs in the
front of the maintenance section... You know, all of that
stuff that nobody reads.

-Chuck Harris

Grayson Evans wrote:

Hello Dave,

I order several BC566's to replace the 2N2207. Do you have any idea what the pin-out is on the 2N2207. I have searched the internet but can't find a data sheet on the 2207. i assume on of the leads is the case, I can't even see the back of the sockets due their being right under the parts on the ceramic strips.
Thanks !

Grayson






Grayson Evans
 

Hello Dave,
I answered my previous questions from earlier today. I found a 2N2207 that I COULD see the connections and traced it out.
So I replaced Q173 and Q184 with BC556 and bingo, fixed the unblanking problem. I haven't paralleled C170 yet, but will do that next.
Thanks for your help and the rest of the guys on this board!
Grayson


Grayson Evans
 

HA! Thanks Chuck, I went through my 547 manual, front to back, and I don't see it anywhere,
but I answered my own questions, I found a 2207 that I could see the socket clear enough and traced it out.
THen I replace the 2207's with BC556's and it works! So i guess I got the pin outs right for a change.
Thanks,
Grayson Evans


Chuck Harris
 

The scoundrels!

I guess they hadn't started that tradition yet when the 545B and 547
were made.

For future reference, there are only a very small handful of ways nonpower
transistors of that era were leaded:

Plastic
-------

---
/EBC\ TO92 Plastic
\.../

---
(ECB) plastic circular bottom with flat side on top section



Metal Can
---------

(EBS C) Metal can w/wo case = S

B
(E C) Metal can w/wo emitter tab
//

B
(E C) Metal can with emitter tab and case = S
//S

And that is it.

Grayson Evans wrote:

HA! Thanks Chuck, I went through my 547 manual, front to back, and I don't see it anywhere,
but I answered my own questions, I found a 2207 that I could see the socket clear enough and traced it out.
THen I replace the 2207's with BC556's and it works! So i guess I got the pin outs right for a change.
Thanks,
Grayson Evans






 

On Sun, Sep 27, 2020 at 01:29 AM, Chuck Harris wrote:


For future reference, there are only a very small handful of ways nonpower
transistors of that era were leaded:

Plastic
-------

---
/EBC\ TO92 Plastic
\.../

---
(ECB) plastic circular bottom with flat side on top section

Metal Can
---------

(EBS C) Metal can w/wo case = S

B
(E C) Metal can w/wo emitter tab
//

B
(E C) Metal can with emitter tab and case = S
//S

And that is it.
Text mode makes layout a bit confusing, on my screen at least, especially for the plastics. One might add the layouts are all bottom view. Some TO-92's have a triangular lead configuration like the metal cans TO-5, TO-39, TO-72.
Bottom/top view distinction is clear for the cases with emitter tabs and irrelevant for the linear arrangements.
By far the most common cases were TO-5, TO-39, TO-72 and TO-92. The flat side of the latter corresponds with the "empty" area on the metal cans' pin layout, leading to the same CBE-arrangement, which is why I considered that the "standard" arrangement.

Raymond


Dave Wise
 

The 2N2207 has a case style that became obsolete later, a metal cylinder with coplanar leads.
Holding the transistor so the leads point down and the ones that are close together on the left, it's E-B--C. Collector is distant from the others.
Some transistors in this case style brought out the case to a fourth lead. Those are E-B-case--C. They shielded C from B quite well, for neutralized common-emitter RF amplifiers.

TO-92: holding it with the flat facing you and leads down, most transistors with 2Nxxx JEDEC numbers are E-B-C, while most BCxxx Pro-Electron's are C-B-E. (And most Asian 2Sxxx are E-C-B. Never a dull moment!)

TO-92, and TO-98 ("plastic circular bottom with flat side on top section") parts always have coplanar leads. Perhaps you (Raymond) are thinking of TO-105 and TO-106, another early plastic package using an epoxy-capped ceramic header which put the leads in a circle a la TO-5 and TO-18, respectively.

TO-92 won the race because it's cheaper to make. Stamped lead frame, attach die, wire bond, jig it into the mold, inject, cut the lead joins and you're done.

Dave Wise
________________________________________
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Raymond Domp Frank via groups.io <hewpatek=gmail.com@groups.io>
Sent: Saturday, September 26, 2020 5:15 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] 547 Repair - how do you locate where parts are located on ceramic strips?

On Sun, Sep 27, 2020 at 01:29 AM, Chuck Harris wrote:


For future reference, there are only a very small handful of ways nonpower
transistors of that era were leaded:

Plastic
-------

---
/EBC\ TO92 Plastic
\.../

---
(ECB) plastic circular bottom with flat side on top section

Metal Can
---------

(EBS C) Metal can w/wo case = S

B
(E C) Metal can w/wo emitter tab
//

B
(E C) Metal can with emitter tab and case = S
//S

And that is it.
Text mode makes layout a bit confusing, on my screen at least, especially for the plastics. One might add the layouts are all bottom view. Some TO-92's have a triangular lead configuration like the metal cans TO-5, TO-39, TO-72.
Bottom/top view distinction is clear for the cases with emitter tabs and irrelevant for the linear arrangements.
By far the most common cases were TO-5, TO-39, TO-72 and TO-92. The flat side of the latter corresponds with the "empty" area on the metal cans' pin layout, leading to the same CBE-arrangement, which is why I considered that the "standard" arrangement.

Raymond


 

On Sun, Sep 27, 2020 at 02:39 AM, Dave Wise wrote:


Perhaps you (Raymond) are thinking of TO-105 and TO-106, another early plastic
package using an epoxy-capped ceramic header which put the leads in a circle a
la TO-5 and TO-18, respectively.
I should have explained that I meant the coplanar TO-92 leads came bent that way in some cases, just outside the package. Of course, being European, we mostly used BCxxx and BFxxx (Pro-Electron) devices. None of that 2N3904 etc. stuff...
We encountered very few TO-105/106 devices, mostly Fairchild but they were uncommon and hardly if ever available in component shops. I'm sure most devices that looked like those (ceramic header with epoxy top) were uL 914 RTL dual NOR gates.... - also Fairchild of course. That was 1965-1970...

Raymond