TDS3014 adventures

[james_s]

Does each channel have its own ADC? There is so little documentation available for these scopes that I actually know very little about the architecture. IIRC they use some sort of CCD technology for the capture memory but I don't know offhand where the separate channels converge.

[giosif]

Well, that is my novice level of understanding & of describing the scope functionality.
Whether those ADG361D ICs are really ADCs or something else, I actually don't know, but those are the ICs I was referring to.
So, since each of the other IC types nearby are in a quantity of less than 4 (to match the number of channels), I suspect the issue might be with one of those, rather than the ADG361D on channel 4 or anything below that ("below" = towards the input terminal).
But this is just me trying to make sense of the behavior I'm seeing, without a solid understanding of scope functionality or inner workings of this particular model.

[giosif]

Today, I had a look at the thermal image inside the scope while it was on and I could see the "ADC" IC for channel 4 (aka U400) stays cold, while all others are getting hot.
So, something is definitely wrong on that channel.
I want to check if U400 at least gets power, but let's see if I can find the right pins (a datasheet would be good now).
I will also ask on the forum for anyone with a spare ADG361D they would like to sell.

[james_s]

Can you find any decoupling capacitors in the vicinity? That's usually a good place to start for finding the voltage to an IC. I'm more inclined to suspect that the chip is not getting enabled rather than not getting power though. Since you're getting data from another channel instead that implies that the same DAC is getting enabled instead of the one you want. It's been a while now since I've poked around inside one of these but look for something like a 74LS138.

[giosif]

There are several caps on the other side of the board (see picture), and only one tantalum on the same side (and doesn't appear to be related to the power rails for the "ADC" ICs).
In terms of ICs in that area of the board other than the ADGxxx ones and processor & memory, I can only see an Elantec 4581CS "sync separator" (quoting from the data sheet, as I don't really know what that means) and an MC33178 op-amp.

As a side note, while investigating the board, I noticed resistor R555 (inside the red rectangle in the attached picture) doesn't have any markings and looks kind of suspicious to me.
Measuring it in-circuit gives me a reading of approx 39 kOhm.
It's a long shot, but anyone here happening to have a similar board at hand that could check the value of that resistor, please?

[james_s]

Sync separator ICs do exactly what the name implies, they take a composite analog video signal and separate out the horizontal and vertical sync signals. In the scope it will be used for the video triggering option, not hugely useful for most people anymore but it was very handy in a former job where I often used a TDS3000 to look at the Macrovision analog copy protection pulses on the analog video outputs of the settop boxes we made. Macrovision used a series of pulses in the blanking interval of several specific video lines to upset the AGC in VCRs to prevent recording protected content and the TDS3000 with that option could trigger on a specific video line.

We also had a few Tek VM700's that were much more capable  in terms of measuring analog video signals however they were limited to 480i standard definition. Not that any of this helps you fix your scope.

I see nothing particularly suspicious about that resistor and 39k is not an unreasonable value for a resistor but I suppose it can't hurt to compare it. I really think the problem is going to be with the address decoding logic that enables the DACs or whatever those are. How that is actually implemented in these I have no idea, it would be fantastic to fully reverse engineer these scopes at some point.

[giosif]

Thank you for the info and apologies for not responding up to now!
I've been waiting for a replacement oscillator I ordered to show up, but over a month later and still no joy.

In the meantime, I've tested the non-working oscillator and it behaved exactly like in your case: it won't do anything for supply voltages above 3.3V but, at about 1.7-2.0V, it would start oscillating at 25.25 MHz.
Afterwards, I decided to have a look inside (see attached pictures).

[james_s]

That's interesting that they both failed in exactly the same way.

While you're waiting, I suspect an oscillator that is approximately the right frequency would probably work ok at least temporarily. Can you find anything in the 70-80MHz range locally? My scope actually booted up with a 27 MHz oscillator, it wasn't usable but something that is at least in the right ballpark might work. I actually have a spare oscillator as I bought two at the time I ordered it, but as I'm in the US I doubt I could get it to you any faster than the one you already ordered. Some mail is very slow these days, I ordered some small parts from Australia that took over 2 months to arrive.

[giosif]

I have already fitted an oscillator that I had - it is running at 50 MHz.
Maybe trying something closer to the 75.75 MHz would be a good test to see if channel 4 would start working properly then.
However, I am not that keen to swap oscillators on the board, in fear I may damage something else.
So, I think I'll wait for the right part to arrive.
Hopefully sooner than 2 months...

[giosif]

Just a quick update: the oscillators I've been waiting for all this time have finally arrived (would have one to spare, in case someone UK/EU-based needs it). 
Installed one in the scope, fired it up and checked channel 4 - the problem persists (signal from channel 3 input replicated on channel 4, no signal from the actual channel 4 input). 

Another issue I had noticed before, but wasn't sure whether it may be due to the wrong frequency of the oscillator (the temporary one was running at 50 MHz) - Ethernet connectivity was not working.
This problem continues to be present even after installing the correct type of oscillator.
Some details: when connecting the Ethernet cable, the link on the switch port comes up (at 10 Mbps, half-duplex), but there is no traffic coming in from the scope (used static IP config on the scope).
I even tried manually entering the MAC address into the switch MAC address table, as well as a manual ARP entry for the IP of the scope, but still no joy.
I checked connectivity between the board with the Ethernet connector and the main board and all seems to be in place.
I'm stating to think whether the main CPU might have some issues...

[james_s]

It's possible of course, but virtually every time I have suspected some expensive, difficult to test and difficult to replace part that has turned out to not be the case. Given the scope mostly works, the CPU itself is one of the last things I'd suspect.

[giosif]

I get what you are saying but, then, for the Ethernet not to be working...
As far as I can tell, the network data goes from the board with the Ethernet connector straight to the main CPU.
Of course, there is the possibility something is broken on the board with the Ethernet connector but, to the degree I could check it (it can be easily powered outside the scope), all seemed ok with it.
(Also, forgot to mention on Ethernet troubleshooting, I tried both straight and crossover UTP cables).

The above said, I will for now treat the two main issues as unrelated and for the fault with the network connectivity as unlikely to be caused by a broken CPU.
Need to think of next steps to troubleshoot the network connectivity problem...

[james_s]

It could be something with the address decoding to select peripherals, that could break ethernet and one of the channels depending on how the address is laid out. It could be a fault in the ROM or RAM, it could be two completely unrelated faults, there are many different things it could be. Is the CPU address bus buffered or do address likes from the CPU connect directly from the CPU itself to the expansion connector? Given the way computers work, a fault in the CPU will normally cause much more serious problems, if an address like is stuck then large swaths of memory and IO will be inaccessible, if any of the lines on the data bus are not functioning then almost nothing will work properly.

[Galen]

My TDS3014B and 3012B, all can't make Ethernet work. Using  crossover cable, The computer said "connected, but I ping the scope, no reply.  The scope looks good for all other functions, only the network, always fail, even I fix the IP at both side. no clue what's wrong.

[madao]

spare parts  TDS3012 is arrivied  for  repair of gosif's  TDS3014B and my  TDS3014.

Sad,   only PSU is dead,   it runs good with  external  15V .  SPC  passed.
I make  think about it...

greetings
matt

[Galen]

spare parts  TDS3012 is arrivied  for  repair of gosif's  TDS3014B and my  TDS3014.

Sad,   only PSU is dead,   it runs good with  external  15V .  SPC  passed.
I make  think about it...

greetings
matt

For the PSU, 3 resisters are very high temperature,especially the 2 serial connected near the transfomer at the board edge.  Does one of them open circuit?

[Galen]

I added a heat sink to the two resisters.
circled the 3 resisters in below picture.

[madao]

your PSU look bit other than my PSU.

But your tip is  good.  Not big resistor in Pictures, small  brigded two resistor is open.  (2x  390 kohm series, your have probably  one 820k, near primary eletrolytics, hidden by  blue ceramic capacitor)
PSU lives again.

Thanks.
Next step: desolder  NVSRAM, battery is empty.

Greetings
matt

[giosif]

Was looking inside the scope again this morning and noticed something I hadn't before: is it just me or the flex cable going from the Ethernet board to the main board is awfully close to those two resistors that are known to get toasty when the scope if turned off (see pictures)?

Now, I have done a continuity check on that flex cable as part of my initial troubleshooting, and all checked ok then, but that was with the cable outside the scope, sitting on a flat surface.
I wonder...

[Galen]

giosif, not only you.   I got 3 TDS3000B recently, the flex cable are all like that. I have to bend this cable to make a little space from the 2 hot resisters.

For the Ethernet, both my two 3000B fail to connect with computer, even I use cross over cable and fix ip at both end. strange. I have no clue.  waiting for your finding.

[Galen]

My third TDS3000B, acquired as dead machine, blank screen, no response to any key.  I guess this is very possible caused by the oscillator. So I bought them. when I measure the out put of the crystal out put, no signal. My normal TDS3012B has 75.757M output. so definitely, this oscillator is dead. then I found it was soldered in wrong direction. check the other oscillator (48M), wrongly installed as well. and both frequency are wrong. 
I bought 2 oscillators and they arrived today. replaced the wrong oscillators. power on the scope, wooo, it boot up normally!
Then I do some measurements. it functins well.
But, it always has a but, I can't find a 75.7575 oscillator, now using a75M, it works well. only the frequency measurement reading is a little wrong, for the 1K square wave, it shows 1.010K.  This is obviously because of the 75M oscillator.
Now I need to find the strange frequency 75.757 oscillator.
BTW, this scope also missed the handle anchor assembly, and the trigger level potentiometer broken. Will try to repair them.
Anyway, already a happy success.  it's time to have a beer.

[AlcidePiR2]

Anyway I desoldered the chip and installed a socket, then popped it in my TL866 and read it as an EEPROM saving the contents to a file.

hi James

Which setting ( which chip) did you use for the reading of the DS1742 on the TL866 ?

Thanks

[madao]

Anyway I desoldered the chip and installed a socket, then popped it in my TL866 and read it as an EEPROM saving the contents to a file.

hi James

Which setting ( which chip) did you use for the reading of the DS1742 on the TL866 ?

Thanks

DS1220 Setting  (both is  2k x8 NVSRAM, which  DS1742 has included  timekeeper.)

[AlcidePiR2]

This is what I did, but  I had an error message when I try to read.

Now I have tried again and realized my mistake.  I had not properly mounted the device. Now everything is fine .

I have attached the .hex file here

I hope this is fine.

I still have two questions :

- Is there a way to know the level of the battery without opening the case ?
- Where can I find a picture of the inside to know where to drill if I want to hook some wires and external battery ?

Thanks

[giosif]

Was looking inside the scope again this morning and noticed something I hadn't before: is it just me or the flex cable going from the Ethernet board to the main board is awfully close to those two resistors that are known to get toasty when the scope if turned off (see pictures)?

Now, I have done a continuity check on that flex cable as part of my initial troubleshooting, and all checked ok then, but that was with the cable outside the scope, sitting on a flat surface.
I wonder...

Just to provide an update here that the Ethernet connectivity issue turned out to be caused by corrupt NVRAM - speaking of which, one way to detect that is if the MAC address of the scope's Ethernet adapter shows in the configuration screen as 08:00:11:01:02:03.
I obtained an NVRAM dump file from another scope and used it on this scope and now network connectivity has been restored. 

Thanks to users YetAnotherTechie and Galen on the forum here for their help sorting this out!