Author Topic: Tek 2246MODA some buttons don't work (Read 24648 times)

RobII · « **on:** October 18, 2015, 05:17:09 am »

Hi All!! New poster so hope this is the right spot. I recently picked up a used Tek 2246 and it mostly works. I am having issues with some of the buttons not working. Specifically, Down buttons for Mode (A, B XY etc), trigger mode, coupling and source. Top menu button, display clear, AC coupling and Channel 3 select all seem to do nothing. The other channels (1,2 and 4) all seem to work ok.

I am new to owning a scope so I need a little direction. I have not yet removed the cover. Is this even worth an attempt to repair?

Thanks in advance!!

Rob

xwarp · « **Reply #1 on:** October 19, 2015, 08:59:26 am »

2246A? or 2246 with Mod A?

I have a 2246A and YES, worth the repair.

Have you downloaded the service manual for it?

http://exodus.poly.edu/~kurt/manuals/uploads/

Get both manuals.

The manual will tell you how to go into service mode and test the front panel buttons.

RobII · « **Reply #2 on:** October 19, 2015, 12:02:05 pm »

I do have both manuals. Problem with getting in to service mode. Manual says to press top and bottom menu buttons, top menu button is not working so can't access service menu. I am debating whether I should open the case.

tec5c · « **Reply #3 on:** October 19, 2015, 12:08:15 pm »

Open it up. It's possible (and likely if the unit has seen some rough days) that the contacts of the buttons have become misaligned/dirty/sticky etc.

Feel free to post some pictures once you've opened it up. Or if you're not confident in opening it up, just say so and we can give you some guidance.

As always, a picture tells a thousand words.

sync · « **Reply #4 on:** October 19, 2015, 12:09:26 pm »

I would carefully reseat the flex cables which go to the front panel first. Maybe it's just a bad contact. My trust in these connectors are limited. I have a 2246A. It's a nice scope.

RobII · « **Reply #5 on:** October 19, 2015, 12:20:34 pm »

Alrighty, I may just take the plunge and give it a go. I will post pics after work today.

Thanks a million!!!

RobII · « **Reply #6 on:** October 19, 2015, 05:25:40 pm »

Sorry xwarp, didn't see your question. It's a 2246MODA.

Quote from: xwarp on October 19, 2015, 08:59:26 am

2246A? or 2246 with Mod A?

I have a 2246A and YES, worth the repair.

Have you downloaded the service manual for it?

http://exodus.poly.edu/~kurt/manuals/uploads/

Get both manuals.

The manual will tell you how to go into service mode and test the front panel buttons.

xwarp · « **Reply #7 on:** October 19, 2015, 06:19:34 pm »

Ok. So 340 of the pdf I am looking at for the Mod A is the switch diagram/interface.

Please list ALL the specific buttons that don't work.

Just want to see if all those that don't work are in common.

Next step will be removing the outer cover and then looking at reseating all the connectors that attach to the front panel.

RobII · « **Reply #8 on:** October 19, 2015, 08:51:38 pm »

Will do. I'm getting out of work in another 10 minutes. I'll get all the buttons and post some pics as well.

Thank you so much!

Rob

Quote from: xwarp on October 19, 2015, 06:19:34 pm

Ok. So 340 of the pdf I am looking at for the Mod A is the switch diagram/interface.

Please list ALL the specific buttons that don't work.

Just want to see if all those that don't work are in common.

Next step will be removing the outer cover and then looking at reseating all the connectors that attach to the front panel.

RobII · « **Reply #9 on:** October 20, 2015, 01:15:30 am »

Here is the list of buttons that do not seem to do anything.

Clear Display
Top Menu button
Fourth Menu button
Volt meter Ch1/2
Channel 3 Mode
Scope BW / 20Mhz
AC Coup Channel 1 and 2
Mode down arrow A, Alt, B, XY
Channel 3 5v / .1v Div
Under the trigger section
Slope
Mode down arrow
Source down arrow
CPLG down arrow

I am attaching pics and circled the affected buttons.

Also the A and B intensity knobs don't seem to do anything either.

xwarp · « **Reply #10 on:** October 20, 2015, 02:14:25 am »

Under the Vertical section, can you select Ch 4?

and then for channel 4, the button that selects the voltage scaling of either 5 or .1 v per/div, does that work?

Check your private messages.

RobII · « **Reply #11 on:** October 20, 2015, 02:22:27 am »

Yes, I can select channel 4 and the 5 /.1 div does come on.

pvt msg received.

Quote from: xwarp on October 20, 2015, 02:14:25 am

Under the Vertical section, can you select Ch 4?

and then for channel 4, the button that selects the voltage scaling of either 5 or .1 v per/div, does that work?

Check your private messages.

tec5c · « **Reply #12 on:** October 20, 2015, 02:35:57 am »

Well that's more faulty buttons than I first thought.

How do you feel about taking the cover off? There's a mesh O-ring like seal around the inside edge of the casing for these models of scopes used for grounding, nothing to worry about now though when you're putting it back together you need to ensure that everything is seated properly and that the mesh is making good contact.

Also, there's a Yahoo group for Tek scopes with an incredible variety of knowledge, perhaps it's worth a browse through there to see if anyone else has had a similar problem.

sync · « **Reply #13 on:** October 20, 2015, 12:21:32 pm »

All switches on the AS3 line of the switch matrix doesn't work. And it looks like that the LEDs on AS3 aren't working too. One trigger mode LED should glow. So the AS3 line is dead.

Have you reseated the front panel connector?
And visually inspect the cable, connector and PCB area for any damage.

tec5c · « **Reply #14 on:** October 20, 2015, 01:22:56 pm »

Well, there you go! That narrows it down by a substantial amount.

Check Q2503 along with all cable connections.

RobII · « **Reply #15 on:** October 21, 2015, 12:09:10 am »

I got the case cover off (NOT FUN!!). Got the potentiometer(sp?) board off. How the hell do I get the switch board assy off and the face plate? I did reseat the ribbon cable here

and tested before I tore into it. No change.

sync · « **Reply #16 on:** October 21, 2015, 12:49:22 am »

No need to disassemble the switch board at the moment. The fault could be a component shown in your photo. Can you check the marked transistor (see attachment)? It drives the AS3 line. You can check it with a multimeter in the diode test function (scope powered off). All 6 possible combinations. Compare it with one or two of the other transistors.

Do you have another scope available?

RobII · « **Reply #17 on:** October 21, 2015, 01:28:30 am »

Awesome. Ok, here's what I get when testing the transistor you pointed out. Forgive me, I'm not sure if these are NPN or PNP so bare with me on my "testing". Checked it with the negative on the middle pin and positive on what was my right side pin.

Got this for a reading with a short beep.

All of the other transistors I tested in the same fashion had this for a reading or slightly higher.

Are we on the right track?

Thanks for your help!!!

xwarp · « **Reply #18 on:** October 21, 2015, 01:37:06 am »

It's a 1 watt Darlington NPN transistor. MPS6725.

In the picture where you see the your leads on the transistor, it's pins from left to right are Emitter, Base, Collector.

RobII · « **Reply #19 on:** October 21, 2015, 01:43:45 am »

So I was testing the Emitter and Base pins. Is that one of the correct ways to test?

xwarp · « **Reply #20 on:** October 21, 2015, 01:50:06 am »

In the picture, the two pins you appear to be on are the base and collector.

If you are up to it, and very carefully, using the negative lead on a ground point, and the positive lead on the square pad you see, which is the Emitter pin, look at the DC voltages and see if that transistor has a different voltage.

RobII · « **Reply #21 on:** October 21, 2015, 01:53:17 am »

I'm assuming you mean power it on and check (i just want to make sure lol).

xwarp · « **Reply #22 on:** October 21, 2015, 02:02:14 am »

Yes, and STAY AWAY from the high voltage supply area.

tec5c · « **Reply #23 on:** October 21, 2015, 02:05:31 am »

Quote from: RobII on October 21, 2015, 01:53:17 am

I'm assuming you mean power it on and check (i just want to make sure lol).

Yeah, with power on.

RobII · « **Reply #24 on:** October 21, 2015, 02:08:43 am »

Ok, From my first pic (pins showing) moving from the bottom transistor to the top

1 = 1.37vdc
2 = .775vdc
3 = .960vdc
4 = .540vdc
5 = 1.60vdc
6 = 1.64vdc

tec5c · « **Reply #25 on:** October 21, 2015, 02:12:25 am »

So 4 = 0.54v is the AS3 line?

RobII · « **Reply #26 on:** October 21, 2015, 02:13:34 am »

Correct, the one you initially pointed me to check.

xwarp · « **Reply #27 on:** October 21, 2015, 02:31:43 am »

Ok, from bottom to top as you have it, and from bottom to top, the transistors are Q2506 to Q2501. Measuring the DC voltage on the square pads, which is the emitter,

Q2501 = 1.5
Q2502 = 1.28
Q2503 = 1.35
Q2504 = .788
Q2505 = .885
Q2506 = 1.34

Now, if you hold the positive lead on the emitter, (square pad), of Q2503, and then press AND HOLD the Channel 1 button, under the vertical mode, you should see the meter drop to .88 vdc.

xwarp · « **Reply #28 on:** October 21, 2015, 02:33:20 am »

One more thing, all of the collectors should have 5 vdc on them, did you check that?

RobII · « **Reply #29 on:** October 21, 2015, 02:47:38 am »

I'm a little confused. Q2503 is the AS3 line and I am only reading .540v on that. I did check the 5v, all are good.

Quote from: xwarp on October 21, 2015, 02:31:43 am

Ok, from bottom to top as you have it, and from bottom to top, the transistors are Q2506 to Q2501. Measuring the DC voltage on the square pads, which is the emitter,

Q2501 = 1.5
Q2502 = 1.28
Q2503 = 1.35
Q2504 = .788
Q2505 = .885
Q2506 = 1.34

Now, if you hold the positive lead on the emitter, (square pad), of Q2503, and then press AND HOLD the Channel 1 button, under the vertical mode, you should see the meter drop to .88 vdc.

xwarp · « **Reply #30 on:** October 21, 2015, 02:51:05 am »

Do me a favor, the readings you took, renumber them to the corresponding transistors that they are for.

Q2501 is closest to the chassis.

Nevermind that....lol....

Something on that line is holding that value down.

It could be that the pads on the switch panel need to be cleaned.

The emitter is pulled from 1.3 vdc down to ~0.8 vdc when Channel 1 mode button is pressed.

RobII · « **Reply #31 on:** October 21, 2015, 02:52:41 am »

So I am not confusing everyone else with my wonderful way of labeling. Based on what you are saying, here are my voltages by the transistor #

Q2506 = 1.37
Q2505 = .775
Q2504 = .960
Q2503 (AS3) = .540
Q2502 = 1.60
Q2501 = 1.64

RobII · « **Reply #32 on:** October 21, 2015, 02:54:52 am »

Got it. Check my previous post. I should have done it that way from the start.

Quote from: xwarp on October 21, 2015, 02:51:05 am

Do me a favor, the readings you took, renumber them to the corresponding transistors that they are for.

Q2501 is closest to the chassis.

xwarp · « **Reply #33 on:** October 21, 2015, 02:58:33 am »

If you decide to pull that panel off and clean it, BE VERY CAREFUL with what you use to clean the carbon traces.

Just distilled water should be good, maybe some diluted windex.

I know that I had issues with a TDS 460 scope and some Fluke 8840's that use carbon for the circuit traces under the button pads.

RobII · « **Reply #34 on:** October 21, 2015, 03:00:20 am »

Will do. So are you thinking it's not a component (transistor) failure?

Quote from: xwarp on October 21, 2015, 02:58:33 am

If you decide to pull that panel off and clean it, BE VERY CAREFUL with what you use to clean the carbon traces.

Just distilled water should be good, maybe some diluted windex.

I know that I had issues with a TDS 460 scope and some Fluke 8840's that use carbon for the circuit traces under the button pads.

RobII · « **Reply #35 on:** October 21, 2015, 03:06:19 am »

LOL, I just saw your edited post. So how do you get the switch panel off? I removed the Pot panel but couldn't figure out how to get the switch panel off.

Quote from: xwarp on October 21, 2015, 02:58:33 am

If you decide to pull that panel off and clean it, BE VERY CAREFUL with what you use to clean the carbon traces.

Just distilled water should be good, maybe some diluted windex.

I know that I had issues with a TDS 460 scope and some Fluke 8840's that use carbon for the circuit traces under the button pads.

xwarp · « **Reply #36 on:** October 21, 2015, 03:10:14 am »

It could be, but based on how dirty that panel is, I'd look at pulling it and cleaning it first.

If you can be careful about it, you could pull that ribbon cable, which is J/P 2501, very carefully bend the number pin out of the way and reinsert the cable.

Then, measure the voltage on the emitter of Q2503. If it reads normal, then your problem is in the panel.

xwarp · « **Reply #37 on:** October 21, 2015, 03:11:56 am »

The front panel looks like it is attached by those black plastic pull pins. You may have to pull the front bezel off before that though.

I've already put my 2246a back together, but can check it tomorrow.

RobII · « **Reply #38 on:** October 21, 2015, 12:05:56 pm »

I'll give it a go. Thanks!!! I appreciate everyone's assistance.

Quote from: xwarp on October 21, 2015, 03:10:14 am

It could be, but based on how dirty that panel is, I'd look at pulling it and cleaning it first.

If you can be careful about it, you could pull that ribbon cable, which is J/P 2501, very carefully bend the number pin out of the way and reinsert the cable.

Then, measure the voltage on the emitter of Q2503. If it reads normal, then your problem is in the panel.

sync · « **Reply #39 on:** October 21, 2015, 03:14:11 pm »

I think Q2503 is dead. Your measurement of 0.3V between base and collector is a strong indication. A silicon junction have about 0.6 - 0.7V forward voltage. Measuring a substantially lower value means a shorted device. There could be a short elsewhere. But this is unlikely.

I would replace Q2503. The availability of MPS6725 is bad. I had a quick look at Digikey and a ZTX603 should work. But there are other possible replacements. Where did you buy your components?

RobII · « **Reply #40 on:** October 21, 2015, 03:29:40 pm »

Thank you for the reply. I was curious as well about that. The reading was so far off I was hoping that might be part of the problem.

As for where I buy components from, I looked at Mouser.com briefly last night and same as Digikey, obsolete. I do see some MPS6725 on Ebay listed as new.

http://www.ebay.com/itm/MPS6725-Motorola-NPN-Darlington-Transitor-50V-1A-TO92-Lot-of-10-/321640691303?hash=item4ae3477667:g:AWsAAOSwzrxUtD7H
or
http://www.ebay.com/itm/MOTOROLA-MPS6725RLRP-One-Watt-Darlington-Transistor-NPN-Silicon-NEW-9-PKG-/190797809949?hash=item2c6c6f891d:m:mfM1fpHFP_GCHSA0t5k9WkQ

Thoughts on those?

xwarp · « **Reply #41 on:** October 21, 2015, 03:48:48 pm »

Quote from: RobII on October 21, 2015, 03:29:40 pm

Thank you for the reply. I was curious as well about that. The reading was so far off I was hoping that might be part of the problem.

As for where I buy components from, I looked at Mouser.com briefly last night and same as Digikey, obsolete. I do see some MPS6725 on Ebay listed as new.

http://www.ebay.com/itm/MPS6725-Motorola-NPN-Darlington-Transitor-50V-1A-TO92-Lot-of-10-/321640691303?hash=item4ae3477667:g:AWsAAOSwzrxUtD7H
or
http://www.ebay.com/itm/MOTOROLA-MPS6725RLRP-One-Watt-Darlington-Transistor-NPN-Silicon-NEW-9-PKG-/190797809949?hash=item2c6c6f891d:m:mfM1fpHFP_GCHSA0t5k9WkQ

Thoughts on those?

As I suggested, before you run off and start easter egging shit, pull that ribbon cable and lift up pin #4 very carefully and reinsert the cable. Then check the voltage on the emitter. At least eliminate the possibility of a stuck button, because guess what!

I turned my scope on and held down one of the buttons that you had circled in your picture, and, guess what, same button issues.

RobII · « **Reply #42 on:** October 21, 2015, 03:51:58 pm »

Trust me, I am not going to do any sort of desoldering etc before I have all input and have tested what has been suggested. Not ignoring your suggestion at all

Quote from: xwarp on October 21, 2015, 03:48:48 pm

As I suggested, before you run off and start easter egging shit, pull that ribbon cable and lift up pin #4 very carefully and reinsert the cable. Then check the voltage on the emitter. At least eliminate the possibility of a stuck button, because guess what!

I turned my scope on and held down one of the buttons that you had circled in your picture, and, guess what, same button issues.

xwarp · « **Reply #43 on:** October 21, 2015, 04:14:41 pm »

I must going blind because I can't find "AS3" signal on the A16 board.

Long night at work, time for bed.

sync · « **Reply #44 on:** October 21, 2015, 04:20:19 pm »

Quote from: RobII on October 21, 2015, 03:29:40 pm

Thank you for the reply. I was curious as well about that. The reading was so far off I was hoping that might be part of the problem.

As for where I buy components from, I looked at Mouser.com briefly last night and same as Digikey, obsolete. I do see some MPS6725 on Ebay listed as new.

http://www.ebay.com/itm/MPS6725-Motorola-NPN-Darlington-Transitor-50V-1A-TO92-Lot-of-10-/321640691303?hash=item4ae3477667:g:AWsAAOSwzrxUtD7H
or
http://www.ebay.com/itm/MOTOROLA-MPS6725RLRP-One-Watt-Darlington-Transistor-NPN-Silicon-NEW-9-PKG-/190797809949?hash=item2c6c6f891d:m:mfM1fpHFP_GCHSA0t5k9WkQ

Thoughts on those?

Normally I would not buy components on Ebay. If your are lucky you get NOS. But there are also fakes. It doesn't have to be a MPS6725. There are a lot of transistors which will work fine in this circuit.

sync · « **Reply #45 on:** October 21, 2015, 04:36:24 pm »

Quote from: xwarp on October 21, 2015, 03:48:48 pm

As I suggested, before you run off and start easter egging shit, pull that ribbon cable and lift up pin #4 very carefully and reinsert the cable. Then check the voltage on the emitter. At least eliminate the possibility of a stuck button, because guess what!

I turned my scope on and held down one of the buttons that you had circled in your picture, and, guess what, same button issues.

Even with a stuck button the LEDs on the AS3 line will work. None of the trigger mode LEDs is lit up on RobII's photo.

sync · « **Reply #46 on:** October 21, 2015, 05:02:19 pm »

Quote from: RobII on October 21, 2015, 03:51:58 pm

Trust me, I am not going to do any sort of desoldering etc before I have all input and have tested what has been suggested. Not ignoring your suggestion at all

If you want to do some measurements then use an oscilloscope. I'm guessing your 2246 will work enough to measure itself. Remove the ground lead from the probe so it doesn't short something accidentally. You don't need the probe ground lead because you are measuring the scope itself. Set the channel to 2V/div, DC coupling and the time base to 2ms/div. Probe the emitters of Q2503 (AS3) and of the other transistors for comparison. This is how the emitter of Q2503 looks on my 2246A. I used a DSO for easier screenshots. It looked the same on my 2246A measuring itself.

RobII · « **Reply #47 on:** October 21, 2015, 05:19:33 pm »

Ah, yes. Understood. Thanks.

Quote from: sync on October 21, 2015, 04:20:19 pm

Normally I would not buy components on Ebay. If your are lucky you get NOS. But there are also fakes. It doesn't have to be a MPS6725. There are a lot of transistors which will work fine in this circuit.

RobII · « **Reply #48 on:** October 21, 2015, 05:42:36 pm »

I will certainly give it a go this evening.

Thank you for your help.

Quote from: sync on October 21, 2015, 05:02:19 pm

If you want to do some measurements then use an oscilloscope. I'm guessing your 2246 will work enough to measure itself. Remove the ground lead from the probe so it doesn't short something accidentally. You don't need the probe ground lead because you are measuring the scope itself. Set the channel to 2V/div, DC coupling and the time base to 2ms/div. Probe the emitters of Q2503 (AS3) and of the other transistors for comparison. This is how the emitter of Q2503 looks on my 2246A. I used a DSO for easier screenshots. It looked the same on my 2246A measuring itself.

xwarp · « **Reply #49 on:** October 21, 2015, 07:35:25 pm »

Not quite accurate:

Quote from: sync on October 21, 2015, 04:36:24 pm

Quote from: xwarp on October 21, 2015, 03:48:48 pm
As I suggested, before you run off and start easter egging shit, pull that ribbon cable and lift up pin #4 very carefully and reinsert the cable. Then check the voltage on the emitter. At least eliminate the possibility of a stuck button, because guess what!

I turned my scope on and held down one of the buttons that you had circled in your picture, and, guess what, same button issues.
Even with a stuck button the LEDs on the AS3 line will work. None of the trigger mode LEDs is lit up on RobII's photo.

RobII · « **Reply #50 on:** October 21, 2015, 07:44:13 pm »

I think that's pretty accurate lol. Proof of concept at the very least. Now to figure out getting the switch panel off. Not sure I'm looking forward to that.

sync · « **Reply #51 on:** October 21, 2015, 07:51:32 pm »

Quote from: xwarp on October 21, 2015, 07:35:25 pm

Not quite accurate:

Sorry, I don't see your point. On RobII's photo none of the trigger mode LEDs are lit up. They are all on AS3. But one of them should glow like shown in your video.

xwarp · « **Reply #52 on:** October 21, 2015, 08:02:49 pm »

o.k., so we know that the emitter of Q2503 is low. We know that when you press a button, the output is pulled low, if you hold a button down on the AS3 line, it will hold that emitter low.

Outside the trigger led's, the other led, (for those buttons) circled, do not change state, just as seen in the video I posted.

Him isolating pin 4 of that cable should eliminate whether it's a stuck button.

I don't know about your 2246a, but mine turns on in the last state that it was in when it was turned off.

We don't know what state his was in and since he can't access the menu to do a reset, he's stuck with where it was last shut down.

xwarp · « **Reply #53 on:** October 21, 2015, 08:08:56 pm »

You know, it could be one of those led's, (in the A/B trigger source line), shorted.

Maybe one of those between DS2025 though DS2032.

ETA:

When you get home, get your meter and on diode function with the positive lead on pin 4 of J2501 cable, probe with the negative leads pins 9-16.

That should at least eliminate those trigger led's.

RobII · « **Reply #54 on:** October 21, 2015, 09:44:04 pm »

Probed with DMM positive on pin 4 and negative to pins 9-16, all had a reading of 1.76.

Pulled pin four and reseated rest of pins, powered on and checked ground point to emitter? (square pad) .31vdc.

xwarp · « **Reply #55 on:** October 21, 2015, 10:27:16 pm »

You've now eliminated the front panel.

RobII · « **Reply #56 on:** October 21, 2015, 10:37:48 pm »

Very cool!! So next step is to get a replacement transistor. I will see about getting the one Sync suggested.

You guys are the best!!!

sync · « **Reply #57 on:** October 21, 2015, 10:51:53 pm »

Do you have any transistors laying around? For testing you can build a replacement with two normal NPNs.

RobII · « **Reply #58 on:** October 22, 2015, 02:21:35 am »

None on hand. I am going to ask one of the repair techs where I work. I am excited about this!!

MarkL · « **Reply #59 on:** October 22, 2015, 02:55:07 pm »

Quote from: sync on October 21, 2015, 03:14:11 pm

I think Q2503 is dead. Your measurement of 0.3V between base and collector is a strong indication. A silicon junction have about 0.6 - 0.7V forward voltage. Measuring a substantially lower value means a shorted device. There could be a short elsewhere. But this is unlikely.

Why would you say that? In circuit, it's entirely possible to have less than 0.7V between the base and collector since that junction is not forward biased. And these are darlingtons, so you're going to see a double drop measuring between base and emitter, so it's not always 0.7V.

RobII, I'm a little confused where you're probing the negative of your DMM in all these measurements, so my apologies in advance if there's a reading that disputes what I'm about to say.

Although it may be Q2503, I think it could also be the output of U2523, the latch that outputs the scan signal. I would check one more thing before you start pulling out transistors.

With your DMM set on *ACV* and negative lead to a ground (chassis is usually good), check the output of U2523 pin 15. That is the output that drives Q2503. You should see some reading there, indicating there's a changing waveform. You can compare this to the other scan outputs on U2523, pins 12, 9, 6, 16, 5. Those should definitely have an AC voltage on them since those buttons are working.

If you have no AC volts on pin 15, or very little (like < 0.1V), switch to DCV mode. If you get a constant voltage close to +5V or 0V, I would say that output on U2523 is dead.

If you do see a comparable AC voltage on pin 15, check to see if you see an AC voltage on the other side of R2536, which is also Q2503's base. You should also see an AC voltage on the emitter of Q2503.

RobII · « **Reply #60 on:** October 22, 2015, 03:50:34 pm »

Thanks for the reply. No worries. I was instructed to have the negative DMM probe on a ground point so I chose a PCB mount screw near the ribbon cable. I also had it on the metal frame to see if there was any difference in my readings. There was not.

I will certainly check what you suggest tonight. Thank you so much for the input!

Quote from: MarkL on October 22, 2015, 02:55:07 pm

RobII, I'm a little confused where you're probing the negative of your DMM in all these measurements, so my apologies in advance if there's a reading that disputes what I'm about to say.

Although it may be Q2503, I think it could also be the output of U2523, the latch that outputs the scan signal. I would check one more thing before you start pulling out transistors.

With your DMM set on *ACV* and negative lead to a ground (chassis is usually good), check the output of U2523 pin 15. That is the output that drives Q2503. You should see some reading there, indicating there's a changing waveform. You can compare this to the other scan outputs on U2523, pins 12, 9, 6, 16, 5. Those should definitely have an AC voltage on them since those buttons are working.

If you have no AC volts on pin 15, or very little (like < 0.1V), switch to DCV mode. If you get a constant voltage close to +5V or 0V, I would say that output on U2523 is dead.

If you do see a comparable AC voltage on pin 15, check to see if you see an AC voltage on the other side of R2536, which is also Q2503's base. You should also see an AC voltage on the emitter of Q2503.

MarkL · « **Reply #61 on:** October 22, 2015, 05:39:21 pm »

Great - let us know what you find.

If nothing obvious is surfacing and Q2503 is still a prime suspect, you could always swap Q2503 with another in the group to see if the problem moves. Since it's only a 3-terminal device it should be easy.

And actually, it's sufficient to just put another in Q2503's place and then check to see if the dead buttons come alive. Some other group of buttons will be dead since there's no driver, but this way it's less soldering if it turns out the original Q2503 was bad. (No sense putting the pads through another round of soldering stress putting in and removing a bad transistor.)

sync · « **Reply #62 on:** October 22, 2015, 06:03:34 pm »

Quote from: MarkL on October 22, 2015, 02:55:07 pm

Quote from: sync on October 21, 2015, 03:14:11 pm
I think Q2503 is dead. Your measurement of 0.3V between base and collector is a strong indication. A silicon junction have about 0.6 - 0.7V forward voltage. Measuring a substantially lower value means a shorted device. There could be a short elsewhere. But this is unlikely.

Why would you say that? In circuit, it's entirely possible to have less than 0.7V between the base and collector since that junction is not forward biased. And these are darlingtons, so you're going to see a double drop measuring between base and emitter, so it's not always 0.7V.

As I written, there could be a short elsewhere but it is unlikely. The base is connected to a 1k resistor and a 4.7nF capacitor. A 1k resistors doesn't turn into a short without very visible damage. The capacitor could cause the short. But it is unlikely. It's not a big MLCC SMD cap. There also could be a short in the PCB. But this is very unlikely on a two layer board.

A silicon darlington has ~1.4V between base and emitter. But when it measures 0.3V then it's dead. (Ok, BIG silicon semiconductors can measure 0.3V with the little current of a DMM. But this is not the case here).

About further measurements. The best would using a scope and look at the waveforms. Then the fault could be narrowed down easily.

sync · « **Reply #63 on:** October 22, 2015, 06:13:40 pm »

Quote from: RobII on October 22, 2015, 02:21:35 am

None on hand. I am going to ask one of the repair techs where I work. I am excited about this!!

They should handle 0.5A or more collector current, eg 2N2222. I measured about 45mA thought one LED. There are 8 LEDs connected on the AS3 line. So this would be about 360mA total (plus some mA if a switch is pressed).

For the final replacement I would go for 1A type like the original.

MarkL · « **Reply #64 on:** October 22, 2015, 06:44:37 pm »

Quote from: sync on October 22, 2015, 06:03:34 pm

A silicon darlington has ~1.4V between base and emitter. But when it measures 0.3V then it's dead. (Ok, BIG silicon semiconductors can measure 0.3V with the little current of a DMM. But this is not the case here).

You can't conclude it's dead. It could simply be off (Vbe < 1.4V).

It's difficult to conclude anything with average DC volt readings taken on dynamic digital signals.

Quote

About further measurements. The best would using a scope and look at the waveforms. Then the fault could be narrowed down easily.

Agree completely. This would take about 2 seconds with a scope, but I'm assuming all we have is the DMM.

EDIT: Fixed quote levels.

RobII · « **Reply #65 on:** October 22, 2015, 06:51:49 pm »

Only scope I have is the partially working 2246 lol. I'll keep everyone posted.

Quote from: MarkL on October 22, 2015, 06:44:37 pm

Agree completely. This would take about 2 seconds with a scope, but I'm assuming all we have is the DMM.

sync · « **Reply #66 on:** October 22, 2015, 07:08:03 pm »

Quote from: MarkL on October 22, 2015, 06:44:37 pm

You can't conclude it's dead. It could simply be off (Vbe < 1.4V).

It's difficult to conclude anything with average DC volt readings taken on dynamic digital signals.

I talking about testing it with the DMM diode function. Not measuring the voltage while the scope is running.

MarkL · « **Reply #67 on:** October 22, 2015, 08:36:53 pm »

Quote from: sync on October 22, 2015, 07:08:03 pm

Quote from: MarkL on October 22, 2015, 06:44:37 pm
You can't conclude it's dead. It could simply be off (Vbe < 1.4V).

It's difficult to conclude anything with average DC volt readings taken on dynamic digital signals.
I talking about testing it with the DMM diode function. Not measuring the voltage while the scope is running.

I'm sorry - you are right. I missed that you were referring to diode test mode.

sync · « **Reply #68 on:** October 22, 2015, 09:54:06 pm »

Quote from: MarkL on October 22, 2015, 08:36:53 pm

I'm sorry - you are right. I missed that you were referring to diode test mode.

I should made it clearer from the beginning. My conclusion that Q2503 is dead is based on the 0.3V of the diode test. The other voltage measurements are difficult to interpret.

RobII · « **Reply #69 on:** October 23, 2015, 02:10:56 am »

Quick edit. I didn't notice there was more than one 74HCT374. Which one is U2523? I checked the one on the right of the pic.

Checked pin 15 (and the others you suggested) all had an AC voltage bouncing around 1.5vac. Also checked resistor R2536 and others near it with voltage about the same. Is it safe to assume transistor Q2503 is the fault or is there something else I should be checking first?

Quote from: MarkL on October 22, 2015, 02:55:07 pm

With your DMM set on *ACV* and negative lead to a ground (chassis is usually good), check the output of U2523 pin 15. That is the output that drives Q2503. You should see some reading there, indicating there's a changing waveform. You can compare this to the other scan outputs on U2523, pins 12, 9, 6, 16, 5. Those should definitely have an AC voltage on them since those buttons are working.

If you have no AC volts on pin 15, or very little (like < 0.1V), switch to DCV mode. If you get a constant voltage close to +5V or 0V, I would say that output on U2523 is dead.

If you do see a comparable AC voltage on pin 15, check to see if you see an AC voltage on the other side of R2536, which is also Q2503's base. You should also see an AC voltage on the emitter of Q2503.

tec5c · « **Reply #70 on:** October 23, 2015, 02:57:01 am »

Do you have a soldering iron? If so, are you comfortable in using it?

As previously suggested, you could swap it with another one of the transistors and see if it fixes the AS3 line.

For such an inexpensive component, I'd just replace it and see if it changes anything. If not, back to the drawing board.

RobII · « **Reply #71 on:** October 23, 2015, 03:02:38 am »

I do and am pretty comfortable using it. I agree about just replacing it. I don't like swapping things around. It just increases the chance of damaging something else.

Quote from: tec5c on October 23, 2015, 02:57:01 am

Do you have a soldering iron? If so, are you comfortable in using it?

As previously suggested, you could swap it with another one of the transistors and see if it fixes the AS3 line.

For such an inexpensive component, I'd just replace it and see if it changes anything. If not, back to the drawing board.

MarkL · « **Reply #72 on:** October 23, 2015, 11:37:58 am »

Quote from: RobII on October 23, 2015, 02:10:56 am

Quick edit. I didn't notice there was more than one 74HCT374. Which one is U2523? I checked the one on the right of the pic.

That's the correct one. The one nearest the connector. (There's a component placement diagram in the service manual, fig 9-9, in the A16 processor board section.)

Quote

Checked pin 15 (and the others you suggested) all had an AC voltage bouncing around 1.5vac. Also checked resistor R2536 and others near it with voltage about the same. Is it safe to assume transistor Q2503 is the fault or is there something else I should be checking first?

I think it was worth confirming that a signal was coming out of pin 15 and getting to Q2503. Latches can go bad, and one or more outputs can get stuck high or low. There are some other bizarre failure modes, but it's probably not the case here.

I would go for Q2503 at this point, as others have already diagnosed.

EDIT: And once you get Q2503 out of circuit, you can re-do the diode test on it that sync described, to make sure it is the culprit. Base to emitter voltage for a good one should be around 1.4V, but you were getting 0.3V in-circuit. Base to collector for a good one should be around 0.7V.

RobII · « **Reply #73 on:** October 24, 2015, 03:14:06 am »

Got the PCB out tonight. Will get Q2503 out tomorrow. I went ahead and placed an order for a couple of replacements.

Thanks for the tip on the component placement. I didn't scroll down far enough in the PDF.

. I will post updates.

Just another quick thanks for all the assistance thus far. You all have been very patient and that has helped so much!!

tec5c · « **Reply #74 on:** October 24, 2015, 03:47:50 am »

I want to see this bit of gear back to its fully functional state!

Thus, if the replacement tranny doesn't fix it, let us know and I'd be happy to help troubleshoot it further.

RobII · « **Reply #75 on:** October 28, 2015, 01:18:26 am »

Had a tech at work replace the tranny with a ZTX603-ND, now there are no lights on the front panel and no buttons work

. I checked DC volts from the square pad on all trannys to ground and 0 volts. Also checked for AC from U2523 and there is AC voltage present with fluctuation. I checked the pin out for the new tranny and I thought was the same as what's installed. Is it possible that the pin out is incorrect and it's in backwards? Would that cause everything else to go dead?

xwarp · « **Reply #76 on:** October 28, 2015, 01:28:14 am »

Go back through and reseat all connectors.

Especially that ribbon cable that goes to the front panel.

Make sure you insert that pin that you isolated a few days back.

RobII · « **Reply #77 on:** October 28, 2015, 01:54:38 am »

Double checked ribbon cables paying attention the one that goes to front panel, no change. Message on the screen "Position Channel 2 DIVS Below Center" Move the position down, and it runs through a self test? Comes up with a warning that the battery may be dead power off and on to verify. Lights are back on but same buttons do not work. Powered off and back on, and it's back to where I started when I got the scope.

RobII · « **Reply #78 on:** October 28, 2015, 01:59:04 am »

Oh, I also noticed that Channel three button is stuck on. That was one that wasn't working before but I can't turn it off now.

RobII · « **Reply #79 on:** October 28, 2015, 12:55:43 pm »

Since this is my first scope, I am still a newb with operations and messages etc. I am gathering that when I removed the ribbon cables, the main board on the bottom loses power from the battery and will go through a self calibration upon restarting. I was not aware of this a couple of posts ago, plus the way I was sitting I didn't see the message on the screen.

Is it possible the replacement tranny I bought is not compatible? Would that cause the channel 3 button to remain on thus killing the rest of the AS3 line buttons? Just spit-balling here.

So, I think I am going to do what someone had suggested, swap a tranny from a working line into the AS3 line and see if there is any signs of life.

Again, thanks for the guidance during this adventure.

tec5c · « **Reply #80 on:** October 28, 2015, 01:21:27 pm »

Do you know what the part number for the original transistor is?

RobII · « **Reply #81 on:** October 28, 2015, 01:26:26 pm »

Quoted from xwarp "It's a 1 watt Darlington NPN transistor. MPS6725."

Quote from: tec5c on October 28, 2015, 01:21:27 pm

Do you know what the part number for the original transistor is?

MarkL · « **Reply #82 on:** October 28, 2015, 02:17:09 pm »

Quote from: RobII on October 28, 2015, 12:55:43 pm

...
Is it possible the replacement tranny I bought is not compatible? Would that cause the channel 3 button to remain on thus killing the rest of the AS3 line buttons? Just spit-balling here.

So, I think I am going to do what someone had suggested, swap a tranny from a working line into the AS3 line and see if there is any signs of life.
...

I had suggested that, but since you have the old Q2503 out of circuit now, you could also do a quick diode test base to emitter, and base to collector on the old Q2503. See if you get the 0.3V base to emitter voltage that you saw in circuit.

And if this is a 2246MODA, the AS3 line is also connected to J2503, if I'm looking at the right schematic. Posted by sync here:

https://www.eevblog.com/forum/repair/tek-2246moda-some-buttons-don%27t-work/msg781558/#msg781558

Where does J2503 go? I'm looking for it, but the PDF version of this manual is excruciatingly slow to load every single page...

sync · « **Reply #83 on:** October 28, 2015, 02:21:26 pm »

Sorry to hear that the scope is still not working.

Quote from: RobII on October 28, 2015, 01:54:38 am

Double checked ribbon cables paying attention the one that goes to front panel, no change. Message on the screen "Position Channel 2 DIVS Below Center" Move the position down, and it runs through a self test? Comes up with a warning that the battery may be dead power off and on to verify. Lights are back on but same buttons do not work. Powered off and back on, and it's back to where I started when I got the scope.

So you didn't change anything and the lights and non-AS3 buttons are working again?

The self calibration and the battery message are caused by loosing/corrupting the non-volatile memory. It's a battery backup SRAM. The battery is on the same board. So removing the connectors doesn't disconnect the battery from the SRAM. Loosing the memory data is not a disaster. It only stores the front panel settings and self calibration data. I think that's the reason that the CH3 LED is now on. You can't toggle it because the switch is on the broken AS3 line. The CH3 LED itself is on the AS0 line. Did you got the message only once or every time you switched the scope on?

If you get it every time then the battery is probably dead. That happen to my 2246A when I turned it on after you started this thread. It now has a new battery. And everything is fine.

Regarding the ZTX603-ND transistor. It is also a 1A, 1W NPN Darlington. It also has the same pin out. Can you make a photo from the replaced transistor?
Do you still have the old transistor?

One possibility for the fault could be that there is a short from the AS3 line to ground. When Q2503 is turned on it will short the +5V power supply to ground and then it will die due overload. This could be the reason that the LED and buttons didn't worked and the memory got corrupted because the +5V power supply where shorted to ground and this crashed the controller.
Can you measure the resistance of the front panel connector (J2501) pin 4 to ground (scope disconnected from mains).

Quote from: RobII on October 28, 2015, 12:55:43 pm

So, I think I am going to do what someone had suggested, swap a tranny from a working line into the AS3 line and see if there is any signs of life.

I wouldn't do that at the moment. If the fault fried the transistors than it will fries the swapped one too.

RobII · « **Reply #84 on:** October 28, 2015, 02:22:43 pm »

I did check the original Q2503 out of circuit and I am pretty sure it was reading open. Negative lead to base correct? It was late last night so I will double check it this evening.

Agreed, the PDF is painfully slow!!

Quote from: MarkL on October 28, 2015, 02:17:09 pm

I had suggested that, but since you have the old Q2503 out of circuit now, you could also do a quick diode test base to emitter, and base to collector on the old Q2503. See if you get the 0.3V base to emitter voltage that you saw in circuit.

And if this is a 2246MODA, the AS3 line is also connected to J2503, if I'm looking at the right schematic. Posted by sync here:

https://www.eevblog.com/forum/repair/tek-2246moda-some-buttons-don%27t-work/msg781558/#msg781558

Where does J2503 go? I'm looking for it, but the PDF version of this manual is excruciatingly slow to load every single page...

RobII · « **Reply #85 on:** October 28, 2015, 02:33:56 pm »

Nothing to be sorry about. It's a learning experience for me so I am up for it.

Originally I didn't see the message on the screen about adjusting the DIV pot down. Once I did that it ran through the self cal then the lights/buttons came back up. Difference now being is the Channel 3 select button is on and I can't turn it off. This is one of the buttons that I believe I had circled back on page one. I did see that battery message one other time and that was after I had removed and reseated the ribbon cables. Since that time I have not seen that message.

I do still have the old transistor and will get a photo of it.

I will check J2501 pin 4 to ground when I get home.

Thank you kindly!!

Quote from: sync on October 28, 2015, 02:21:26 pm

Sorry to hear that the scope is still not working.

So you didn't change anything and the lights and non-AS3 buttons are working again?

The self calibration and the battery message are caused by loosing/corrupting the non-volatile memory. It's a battery backup SRAM. The battery is on the same board. So removing the connectors doesn't disconnect the battery from the SRAM. Loosing the memory data is not a disaster. It only stores the front panel settings and self calibration data. I think that's the reason that the CH3 LED is now on. You can't toggle it because the switch is on the broken AS3 line. The CH3 LED itself is on the AS0 line. Did you got the message only once or every time you switched the scope on?

If you get it every time then the battery is probably dead. That happen to my 2246A when I turned it on after you started this thread. It now has a new battery. And everything is fine.

Regarding the ZTX603-ND transistor. It is also a 1A, 1W NPN Darlington. It also has the same pin out. Can you make a photo from the replaced transistor?
Do you still have the old transistor?

One possibility for the fault could be that there is a short from the AS3 line to ground. When Q2503 is turned on it will short the +5V power supply to ground and then it will die due overload. This could be the reason that the LED and buttons didn't worked and the memory got corrupted because the +5V power supply where shorted to ground and this crashed the controller.
Can you measure the resistance of the front panel connector (J2501) pin 4 to ground (scope disconnected from mains).

Quote from: RobII on October 28, 2015, 12:55:43 pm
So, I think I am going to do what someone had suggested, swap a tranny from a working line into the AS3 line and see if there is any signs of life.
I wouldn't do that at the moment. If the fault fried the transistors than it will fries the swapped one too.

sync · « **Reply #86 on:** October 28, 2015, 03:02:06 pm »

Quote from: MarkL on October 28, 2015, 02:17:09 pm

Where does J2503 go? I'm looking for it, but the PDF version of this manual is excruciatingly slow to load every single page...

This is J2503. Nothing connected.

MarkL · « **Reply #87 on:** October 28, 2015, 10:16:52 pm »

Quote from: RobII on October 28, 2015, 02:22:43 pm

I did check the original Q2503 out of circuit and I am pretty sure it was reading open. Negative lead to base correct? It was late last night so I will double check it this evening.

Positive lead to base for NPN. Base to Emitter should be 1.4V, base to collector 0.7V.

Backing up and double checking what works and what doesn't, I didn't see that you reported "Channel 1 Vertical Mode" as not working. It's also on the AS3 line. Is it non-functional too? I would expect so.

You did report that "Slope" was not working. Correct? That's on AS4.

And confirming: A (or B) intensity still doesn't work?

Quote

Agreed, the PDF is painfully slow!!

I was able to have ghostscript re-assemble the service manual with this command:

Code: [Select]

gs -sDEVICE=pdfwrite -sOutputFile=2246_new.pdf -dSAFER -dNOPAUSE -dBATCH -r150 2246_orig.pdf

I don't know what the issue is with the original, but this allows each page to load at least 10x faster.

MarkL · « **Reply #88 on:** October 28, 2015, 10:24:41 pm »

Quote from: sync on October 28, 2015, 02:21:26 pm

One possibility for the fault could be that there is a short from the AS3 line to ground. When Q2503 is turned on it will short the +5V power supply to ground and then it will die due overload. This could be the reason that the LED and buttons didn't worked and the memory got corrupted because the +5V power supply where shorted to ground and this crashed the controller.
Can you measure the resistance of the front panel connector (J2501) pin 4 to ground (scope disconnected from mains).

Did he not already measure the voltage of AS3 w/ respect to ground here? :

https://www.eevblog.com/forum/repair/tek-2246moda-some-buttons-don%27t-work/msg781959/#msg781959

But perhaps there's a short on the other side of CR2003 on the front panel?

Rob: When doing the resistance check, can you also do a diode test from J2501 pin 4 to ground? (DMM negative to ground.) And then compare with some of the other AS lines?

sync · « **Reply #89 on:** October 29, 2015, 12:07:01 am »

Quote from: MarkL on October 28, 2015, 10:16:52 pm

You did report that "Slope" was not working. Correct? That's on AS4.

The Slope LED is on AS3. Perhaps the switch is working but Rob can't see it.

Quote from: MarkL on October 28, 2015, 10:24:41 pm

Did he not already measure the voltage of AS3 w/ respect to ground here? :

https://www.eevblog.com/forum/repair/tek-2246moda-some-buttons-don%27t-work/msg781959/#msg781959

But perhaps there's a short on the other side of CR2003 on the front panel?

Yes, he did and the 0.54V didn't look like a short. But I just want to rule out a short.
A short after CR2003 is an interesting idea. That could explain it.

btw: Thanks for the ghostscript tip!

But I really want to see signal waveforms. That would be much more productive.
Rob: Is your scope able to display waveforms?
Do you have probes?

RobII · « **Reply #90 on:** October 29, 2015, 01:22:42 am »

Checked the original tran. on diode mode, Base to Emitter = 1.244, Base to Collector = .682. Check for short from J2501 pin 4. No short on diode mode. Checked other pins, pin 7, 8 and 17 short short to ground on J2501.

A and B intensity was working (the knobs were not seated properly on the pots) BUT since last night there is no "sweep"? What I'm trying to say is before the diode change I could see the flat line of the channels (1,2 and 4) but now there is nothing but I see a dot on the screen that sweeps left to right so something has gotten worse.
Prior to the swap I could use channel 1, 2 or 4 and get a wave form, now nothing.

@sync, I do have probes but as it is now, it doesn't display any waveform...or anything for that matter.

The dot is in between

Buttons (sorry for the blurry pic)

Note the Scope BW button is stuck on as well as the Channel 3 button

sync · « **Reply #91 on:** October 29, 2015, 02:11:15 am »

Quote from: RobII on October 29, 2015, 01:22:42 am

Checked the original tran. on diode mode, Base to Emitter = 1.244, Base to Collector = .682. Check for short from J2501 pin 4. No short on diode mode. Checked other pins, pin 7, 8 and 17 short short to ground on J2501.

OK, it looks like that the transistor is ok. So it's not the culprit.

Quote

A and B intensity was working (the knobs were not seated properly on the pots) BUT since last night there is no "sweep"? What I'm trying to say is before the diode change I could see the flat line of the channels (1,2 and 4) but now there is nothing but I see a dot on the screen that sweeps left to right so something has gotten worse.
Prior to the swap I could use channel 1, 2 or 4 and get a wave form, now nothing.

The scope lost it's settings (battery/memory error message). Your picture show a time base setting of .5s (bottom right of the screen). So the dot should move 1 division per 0.5s. That's slow. Turn the time base control (A and B SEC/DIV) clockwise to get faster sweep speeds. Then you should see a line again.

Set the time base to 5ms. Set CH1 VOLTS/DIV to 20mV. It's the most left value at the bottom of the screen. Set the probe to 10x and connect it to the "CALIBRATION" output (left of the CH1 input BNC). You should see a square wave.

RobII · « **Reply #92 on:** October 29, 2015, 02:28:33 am »

YES!! Excellent. Yes, I see a square wave now for channel 1!!!

sync · « **Reply #93 on:** October 29, 2015, 02:22:02 pm »

Good!

Now it's time to measure the faulty circuit. Remove the ground lead from the probe to avoid shorting something accidentally with it. You don't need it because the scope shares the ground internally.

Set the time base to 2ms. Set CH1 VOLTS/DIV to 0.2V which is really 2V/division using a 10x probe. Measure pin 5 of J2501 this is the working AS4 line. You should see a waveform like this.

Important are the positive rectangular pulses. This is when the transistor is switch on. Now probe pin 4. This is the AS3 line.

If you see also these positive pulses then AS3 line on the processor board is good and the fault should be on the front panel switch board.

If not then probe the base of Q2503 and pin 6 of U2523 (74HCT374). There should be also positive pulses.

Please make photos of the waveforms so we can see them.

MarkL · « **Reply #94 on:** October 29, 2015, 03:49:35 pm »

Ok, so the old Q2503 was ok. Time to go back and look at the readings with this knowledge.

The 0.3V reading from early on is bothering me. From Q2503 base (DMM-) to collector (DMM+) was showing 0.3V:

https://www.eevblog.com/forum/repair/tek-2246moda-some-buttons-don%27t-work/msg781931/#msg781931

This is in contrast to the other transistors measured in the same way at 1.0V or so.

Because Q2503 is ok, the diode test current is not flowing through Q2503 in that shot, so the only other possibilities are R2536 and then to U2523 pin 6, or through C2523.

Could be a leaky C2523 (100V ceramic, so not high on my list), or something wrong with the output drivers in U2523.

Rob: Back a couple of pages I told you the wrong pin for the ACV measurements for U2523. It is pin 6 controlling Q2503 and not 15 (too many parallel lines - sorry about that). I was suspicious of U2523 and I might have taken us off track because of this mistake.

If you want to continue with the DMM, you could do a diode test on pin 6 and for comparison a couple of the known good output pins on U2523 (12, 9, 15, 16, 5). Test in both directions. I think pin 6 is going to measure differently. (Scope powered off, obviously.)

I'm only suggesting it because you might be more comfortable with the DMM. On the other hand, troubleshooting a broken scope with itself is certainly an interesting way to learn.

RobII · « **Reply #95 on:** October 30, 2015, 10:03:35 pm »

Sorry again for the delay everyone.

OK, Pin 5 (AS4 Line) Looks good.

Pin 4 (AS3 Line) Not so good.

Base of Q2503. Not so good.

Pin 6 of U2523. Not so good.

Now here's where it get's interesting. While trying to get the probe steady on the base of Q2503 it accidentally touched and shorted base to the collector pin just briefly. For several seconds after that, the buttons that were not working, were suddenly working but only for a few seconds. Eventually they went dead.

Does this say anything about U2523 or maybe a component in between that and Q2503?

RobII · « **Reply #96 on:** October 30, 2015, 11:34:12 pm »

I checked R2536 in circuit - 1000ohms, same for the others. I don't have an capacitance meter so I set my Fluke to ohms and checked C2523 in circuit , 463ohms. The others (C2522, C2524, C2525 and C2526) all were reading 2.9 mohms and 3 mohms.

No worries on the giving the wrong pin. Completely understandable.

Quote from: MarkL on October 29, 2015, 03:49:35 pm

Because Q2503 is ok, the diode test current is not flowing through Q2503 in that shot, so the only other possibilities are R2536 and then to U2523 pin 6, or through C2523.

Could be a leaky C2523 (100V ceramic, so not high on my list), or something wrong with the output drivers in U2523.

Rob: Back a couple of pages I told you the wrong pin for the ACV measurements for U2523. It is pin 6 controlling Q2503 and not 15 (too many parallel lines - sorry about that). I was suspicious of U2523 and I might have taken us off track because of this mistake.

If you want to continue with the DMM, you could do a diode test on pin 6 and for comparison a couple of the known good output pins on U2523 (12, 9, 15, 16, 5). Test in both directions. I think pin 6 is going to measure differently. (Scope powered off, obviously.)

I'm only suggesting it because you might be more comfortable with the DMM. On the other hand, troubleshooting a broken scope with itself is certainly an interesting way to learn.

MarkL · « **Reply #97 on:** October 31, 2015, 12:15:33 am »

Great job getting the traces up!

Pin 6 of U2523 looks happy to me, although admittedly I wasn't expecting that. I've encountered a lot of dead latches. But yours has a full 0 to 5V swing on the output pulse. What doesn't look good to you?

When you shorted the collector to base, you applied +5V to the base and fully turned on the transistor and caused the AS3 line to go high. But since this was not synchronized with the scan, lots of things probably came alive as well as the dark LEDs. At the same time it also probably caused a lot of current to flow into the base. Not sure if this did any damage.

Is it back the way it was now?

~~Can you indicate where 0V is on the shots of Q2503 base and emitter (AS3)? If you can control the vertical position, you could position 0V on the center horizontal graticule line.~~ (Never mind - you probably didn't move vertical position between shots.) At the moment, the voltage swings appear inconsistent with eachother and also, the voltage on the emitter is decaying more slowly than your shot of AS4.

I think you looked before, but you're sure Q2503 collector has a solid +5V on it?

Earlier, xwarp had you isolate AS3 by bending pin 4. It might be useful to look at another screen shot of Q2503 emitter and base, but this time isolated from the front panel.

RobII · « **Reply #98 on:** October 31, 2015, 01:11:39 am »

Thanks but that was all thanks to sync for the excellent guidance getting the traces back.

Oh, U2523 pin 6 looks good then? From what sync had said I was expecting more of what pin 5 (AS4 line) looked like for the ribbon cable.

When I shorted the base and collector not all lights came on but certain ones did and the buttons that were not functional started working (AC Coup, Display Clear, Down select for the mode) but they did not stay working. After a few seconds some of the lit LEDs (ones on the AS3 Line) starting to blink and went dead. So it is back to no buttons on the AS3 line working.

Correct, I did not move the vertical position between probing Q2503 and pin 6 U2523.

I did check a couple of times to make sure there was +5v on the collector for Q2503 and all along that rail had a solid +5

Ok, I will try to isolate the pin again but I don't want to do this too many more times as I'm afraid I'll stress it and snap it off lol.

sync · « **Reply #99 on:** October 31, 2015, 01:23:47 am »

Thanks for the screenshots. They help a lot. MarkL has already analyzed them.

Quote from: RobII on October 30, 2015, 11:34:12 pm

I checked R2536 in circuit - 1000ohms, same for the others. I don't have an capacitance meter so I set my Fluke to ohms and checked C2523 in circuit , 463ohms. The others (C2522, C2524, C2525 and C2526) all were reading 2.9 mohms and 3 mohms.

I think C2523 is shorted/very leaky. That would explain the 0.3V from the in circuit diode test of Q2503. A leaky C2523, which behaves like a low ohm resistor, would build a voltage divider with R2536. That would pull down the base voltage. Like in the 3rd screenshot. And Q2503 isn't switch on anymore.

MarkL · « **Reply #100 on:** October 31, 2015, 02:44:58 am »

Quote from: sync on October 31, 2015, 01:23:47 am

Thanks for the screenshots. They help a lot. MarkL has already analyzed them.

Quote from: RobII on October 30, 2015, 11:34:12 pm
I checked R2536 in circuit - 1000ohms, same for the others. I don't have an capacitance meter so I set my Fluke to ohms and checked C2523 in circuit , 463ohms. The others (C2522, C2524, C2525 and C2526) all were reading 2.9 mohms and 3 mohms.
I think C2523 is shorted/very leaky. That would explain the 0.3V from the in circuit diode test of Q2503. A leaky C2523, which behaves like a low ohm resistor, would build a voltage divider with R2536. That would pull down the base voltage. Like in the 3rd screenshot. And Q2503 isn't switch on anymore.

I think I agree at this point. But I've been hesitating to conclude on C2523 mainly because I don't think I've ever seen a ceramic fail like that in a low voltage circuit.

The DMM resistance readings could have gotten thrown off by a high test voltage on ohms that was turning on CR2003 with a short on the other side on the front panel board. Or perhaps there's something else happening on the front panel board that we haven't thought of yet. Hence, the request to re-do the scope traces with pin 4 disconnected.

The emitter would be floating at that point, so it might look pretty strange, but if the base did not return to a full 0 to 5V swing, there's really nothing else it could be except C2523. (Barring some really strange problem with the PCB and/or traces.)

By my own logic a couple posts back it has to be C2523. But I've already made a couple of mistakes in this thread and I wanted to be sure before popping more components off the board.

If the pin 4 test is not sounding attractive, Rob, an alternative would be to lift one side of C2523 and measure ohms on it. I guess pick whichever one you dislike the least. Depending on how critical C2523 is to the circuit, the scope might even work at that point.

sync · « **Reply #101 on:** October 31, 2015, 02:24:15 pm »

Quote from: MarkL on October 31, 2015, 02:44:58 am

I think I agree at this point. But I've been hesitating to conclude on C2523 mainly because I don't think I've ever seen a ceramic fail like that in a low voltage circuit.

I also found a faulty C2523 unlikely. For the same reason. I never saw such a ceramic cap failing. Only a few higher value MLCC SMD caps. But the measurements and waveforms matches a failed C2523 so well. Of course it could be a even more obscure fault.

Quote

If the pin 4 test is not sounding attractive, Rob, an alternative would be to lift one side of C2523 and measure ohms on it. I guess pick whichever one you dislike the least. Depending on how critical C2523 is to the circuit, the scope might even work at that point.

The circuit will work without C2523. If it's the culprit then the scope should be working after removing it/lift one side.

MarkL · « **Reply #102 on:** October 31, 2015, 03:21:46 pm »

Quote from: sync on October 31, 2015, 02:24:15 pm

The circuit will work without C2523. If it's the culprit then the scope should be working after removing it/lift one side.

And I'll add, even if the scope works after disconnecting C2523, it should still be replaced.

C2523 and R2536 (and the other RC pairs) slow down the edge for the AS lines going to the front panel, probably to reduce EMI eminating from the front panel.

RobII · « **Reply #103 on:** October 31, 2015, 06:28:26 pm »

A quick update. Before I go lifting C2523 I wanted to isolate pin 4 from the ribbon cable and the ohm reading C2523 was the same. Checked with the scope base of Q2503 and it has the same reading as with pin 4 inline, same for pin 6 of U2523. I will lift out a leg of C2523 and report back. Stay tuned and thanks again!!

RobII · « **Reply #104 on:** October 31, 2015, 07:50:44 pm »

As Prof. Farnsworth from Futurama would say, GOOD NEWS EVERYONE!! It lives! Lifted out a leg of C2523,

Powered on and ALL the buttons that were not working, are now working!!!!

All four channels on,

All channel buttons on,

Channel 3 (previously not working)

so to recap, I am now able to Select channel 3, select AC coupling channel 1 and 2, Scope BW, down buttons for Mode, trigger mode, source and CPLG, Slope, 5v for channel 3, Clear Display and Voltmeter.

I do plan on replacing C2523 but I want to ask an opinion on a different "issue". After the scope has been pn for about a minute, it develops a high squeal. Not terribly loud (to me at least) but noticeable. I have experience working with arcade CRT chassis and that squeal usually indicates a faulty flyback. Is it worth pursuing repairing that or just live with it considering I won't be using this scope 24/7.

I can't thank you guys enough for the guidance and patience. Now, does anyone here know anything about troublshooting an Asteroids PCB

RobII · « **Reply #105 on:** October 31, 2015, 08:12:52 pm »

Also, will this cap work as a replacement?

http://www.digikey.com/product-detail/en/C410C472K1R5TA7200/399-4475-1-ND/818332

MarkL · « **Reply #106 on:** October 31, 2015, 08:51:37 pm »

"Eeeexcellent!", in a Mr. Burns voice. Congrats on the repair!

Leaky ceramics are a new one for me.

On the replacement cap, the Tektronix manual lists AVX SA101C472KAA. Mouser carries the exact replacement, but the part you've picked out from DigiKey is the equivalent from a different manufacturer and is perfectly fine.

Squealing after a few minutes is usually from power supply components heating up, and the mechanical expansion frees pieces to move with the switching frequency. I've had it happen with transformer windings most often, but I've also heard stories of it happening with plates inside capacitors. Not different at all from noisy flybacks.

If it bothers you, I would take a well insulated stick and put light pressure on the main switching transformer in different places, or on other areas where you think the sound might be coming from. It's possible the sound is coming from a location deep inside a part and there's nothing you can do about it.

I can tell you from experience that over the years high-pitched sounds will reduce in intensity (hint: I'm *not* talking about the equipment).

Asteroids was one of my favorite games. I enjoyed the 4-2 option game. A 6502 system shouldn't be too hard to fix. You should start a new thread for that one and I'm sure there's a bunch more people who would dive in. With these types of microprocessor systems (6502, 6800, Z80, etc.), it's usually handy to have a logic analyzer too.

RobII · « **Reply #107 on:** October 31, 2015, 09:06:50 pm »

Order placed from Mouser!

I have tinnitus so the squealing doesn't really bother me. It was bothering my wife and daughter as I was troubleshooting but since I won't be using it in their presence, I'll let it go.

I'll start a new thread for my Asteroids PCB. The game plays blind so I needed the scope so I could check the vector output etc.

It's a good day today!!

tec5c · « **Reply #108 on:** November 01, 2015, 07:32:41 am »

Nice to see the repair completed. Good work.

sync · « **Reply #109 on:** November 01, 2015, 06:42:46 pm »

Congratulation! I'm glad it's working again.

Quote from: MarkL on October 31, 2015, 08:51:37 pm

Squealing after a few minutes is usually from power supply components heating up, and the mechanical expansion frees pieces to move with the switching frequency.

Just an idea about the squealing. Does the fan runs? It's not quiet so you should hear it.

RobII · « **Reply #110 on:** November 01, 2015, 08:29:55 pm »

All credit goes to you guys. I merely followed your instructions.

@sync yes the fan does work.

RobII · « **Reply #111 on:** November 07, 2015, 01:07:46 pm »

One final update to close this out.

Old cap out.

Side by side old vs new

New cap in.

I let it run for about 30 minutes and she is still chugging along just fine.

I figured out why the case was so hard to remove, it appears to either have been dropped or had something impact one side. The metal was pushed in by the handle and on the bottom where the plastic feet are, same deal. Got the rubber mallet out and now the case slides on and off very easily.

Once again. Thanks to you guys for the patience, guidance and knowledge!!

REPAIR COMPLETED!!!


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee