TDS 544A Maybe Bad Flyback. (Crack Found) Need NVRAM Dumps..

[KG7AMV]

Turn it on odd the screen is shrunken.. 



Then went dim and now sounds like high voltage arc left hand corner? Thinking flyback?

I opened it up and disconnected the IDC for the CRT Board so I know it's in that can! Left Rear where she be arching..

I will open her up further tomorrow.

I know its due or way past due for the aluminum electrolytic caps to be replaced they will go on my arrow order next week Unless anyone has a list of them I am seeing 33uf 10V, 10uf 35V, and 3.3uf 50V



 

 

[CaptDon]

The 'A' series will almost surely need the caps changed and the
flybacks were known to go bad. You may be lucky when pulling
out the display board you will find a bad solder connection around
the flyback on the underside of the circuit board. The Nu-Scope
LCD upgrade seems nice, but I feel very overpriced. You can find
CRT driver boards on Ebay but beware of Chinese sellers. If they
sell you junk you have to pay to send it back which is more than
the purchase cost and then they won't refund your money anyway.
Parts from Test-Key are so overpriced it is nearly criminal!!!

[CaptDon]

Most of the Acquisition boards also need washed in addition to changing
the caps! Don't know what works best (unless you work at a board house
and have access to a nice de-ionized wash system) but I used hot soapy
water (Dawn) and a very soft brush then flushed with clear warm water
and then spent about 1/2 hour with a variable heat variable speed hair
dryer. My Aq board was failing even after the cap change and had some
really weird artifacts in the waveform. After the bath it passed SPC and
works fine to this day!! Be Careful De-Soldering!!! The electrolyte has
damaged the traces and they will tear/decompose easily. The smoke
smells like dead fish and oxidized rubber.
 

[madao]

Grinding and drilling on crack.  Then coat it with 2k glue.

I have repair  TDS744  flyback transformer with this methode.   It run stable.

Important: desolder all bad SMD eletrolytics, whatever it look good.  Washing of PCB is very neccesary, or  it break sometime down.  (i use dish wash detergent  & water)

Greetings
matt

[james_s]

A shrinking display can be caused by the EHT rising. It's possible that another failure caused excessive HV which shot through the insulation of the flyback and arced to the core.

The Dallas chips can be problematic to read in many programmers, but you shouldn't need to mess with those right now, IIRC the scope will boot up and work without at least one of them even there, aside from the clock they are used to store settings, reference waveforms, enabled options and other non-critical stuff.

[Rollin Hand]

I had a tv shop and I repair some fly back this way ,they didn't last too much 1-2 years,better procure another flyback.

[CaptDon]

I hope you did the 'minimalfloppydump' to a floppy disk before you pulled
those Dallas chips!!! The data is 99.9% guaranteed to be scrambled no
matter how carefully you remove them!!!! I highly recommend and stand
behind the Ebay seller who sells the replacement set for $100.00 for the
1650/1250 and 1486 pair!! He has the restore files on hand and can pre-
load them for you if your data is toast. He also supplies the sockets for
your new chips. I did a TDS644A and a 644B with his parts and was highly
satisfied. All of my data was scrambled upon desoldering!! I read the data
with both a ChipProg48 and a BP Microsystems unit and even just for the
heck of it re-wrote good data into the old scrambled Dallas chips to see
what would happen. I did have success to both read and write them. I have
data dumps of my 644A, 644B and the company 654C.

[james_s]

I must have been very lucky because a few years ago I removed the Dallas chips from my TDS784C, dumped them and installed low profile sockets (video trigger board won't fit with standard sockets) and the data remained intact. As far as I know there is nothing super critical stored in those though, the SPC constants are but you should run SPC periodically anyway. I think the important calibration stuff is in EEPROM on the acquisition board.


Regarding patching up flybacks, I have only one data point but when I was a teenager I found a 25" Sharp TV by the dumpster at the apartment complex where we lived. I dragged it home and found a bad capacitor next to the flyback and then the flyback was arcing to the frame. I glopped a bunch of hot melt glue over the crack and it worked fine for years, it was still working when I gave away the TV several years later.

Looking at that photo of the screen again, the fact that the image appears very bright and somewhat washed out also suggests the EHT has gone way high, I'm not familiar with how it is regulated in these monitors, thankfully I've never had to try to fix mine. Given the unusual NuColor display runs at a high field rate I'm wondering if the HV is separate circuit independent of the horizontal scan? If that's the case it might not be too hard to adapt it to use some other flyback transformer. I do love those NuColor displays, it's such a fascinating and unusual technology and it produces a very unique look, a sharp clear image as produced by a monochrome CRT with no screendoor effect of a shadow mask, yet in full color. There was only one other application I'm aware of where it was ever used, in a small studio broadcast monitor, I'd love to find one of those but they're quite rare.

[CaptDon]

The poster on these forums named FLYTE may have those backups.
I am almost positive the Ebay seller of the newly designed replacement
pair 1250/1650 1486 will be able to help you. The first 2k or 4K (depending
on zero ohm jumpers) of your NVRAM is a protected area with some
data that was set at the factory. If you try to insert blank or scrambled
NVRAM's it will be a nightmare. Data dumps from matching models are
close enough to get you up and going. There are even hacks listed to
enable features that are not found on your scope although the physical
hardware must be there to support the feature e.g. 'T.V. Sync' module.

[Rollin Hand]

TV Trigger board

My tds724d has a trigger board and on the startup it tell me option 05 fail.
And in the error page it said :
opt TV dont care codes
SW to n trig2 no output
Is the tv trigger board bad? I check all the power voltages and they are ok .The capacitors on this  board are ok too.
Thanks for the help
Can I remove this option without removing the tv trigger board true GPIB?

[james_s]

I got that error once when one of the connectors was not fully seated. It's pretty easy to send the GPIB command to disable that option or enable other features. You can build a GPIB adapter for under $10 so there's no reason not to have one.

[james_s]

I built a set for mine, the work great.

Make sure you use low profile sockets if you have or think you might ever want the video trigger board, there is very little clearance.

[CaptDon]

We had a TDS540A at work. It was sent to the Cal Lab for its yearly Cal.
The morons in Cal lab couldn't get something to cal on it so they threw it
in the dumpster. I didn't find out for several weeks. I sent an email over
"Hey, what happened to our 540A?" the reply "We couldn't get it to cal
so we scrapped it". Those freaking morons!!!! My co-worker and I were
using it with perfect results right up to when they took it. I could have
used the boards!!! Our Cal guys are numb nuts anyway, they probably
did something stupid in their setup. I swear the scope was working fine!!

[james_s]

Wow that's aggravating.

The last place I worked that did hardware used to send stuff out to be calibrated regularly and several times gear that had been working perfectly came back with problems, screws missing, things not working right, but hey it had a valid cal sticker.

[andy2000]

I must have been very lucky because a few years ago I removed the Dallas chips from my TDS784C, dumped them and installed low profile sockets (video trigger board won't fit with standard sockets) and the data remained intact. As far as I know there is nothing super critical stored in those though, the SPC constants are but you should run SPC periodically anyway. I think the important calibration stuff is in EEPROM on the acquisition board.

I've never had any trouble removing and reading Dallas NVRAM chips without corrupting them, but I'm always extra careful.  You can tie the write protect pin to to VCC for extra protection when reading. 

The 540A and 544A don't have EEPROMs on the acquisition board.  The calibration is stored in the DS1650, and should be backed up.  It's not a bad idea to copy the backup to an SD card and tape it to the inside of the scope.

The clock chip won't verify because the time is constantly changing, but the NVRAM should. 

[james_s]

I've never had any trouble removing and reading Dallas NVRAM chips without corrupting them, but I'm always extra careful.  You can tie the write protect pin to to VCC for extra protection when reading. 

The 540A and 544A don't have EEPROMs on the acquisition board.  The calibration is stored in the DS1650, and should be backed up.  It's not a bad idea to copy the backup to an SD card and tape it to the inside of the scope.

The clock chip won't verify because the time is constantly changing, but the NVRAM should.

I corrupted the calibration on more than one DS1644(?) in a TDS300 series doing that so it is possible at least with those, maybe when the battery is extremely low? I don't know.

Hm that's good to know, when did they move to EEPROM for the cal data?

[Rollin Hand]

Each time I read the nvmas is a difference in 5-8 bytes on each reading,in different places.Which reading is a good one?
I did the reading 6 times.

[james_s]

What are you reading it with? I had trouble reading many different NVRAMs in my TL866, some would read wrong the first time and correctly immediately after. I suspect it's caused by power to the chip being applied immediately before reading.

[CaptDon]

The 1486 will never verify twice because the real time clock
shows up as normal ram locations within the chip and are
always changing.
As for corrupting the Dallas DS chips when removing, these
chips on the older scopes have code dates all the way back
to the early 90's and I haven't had one yet that had usable
data after desoldering. Usually just endless strings of 55 55 55
AA AA AA and random garbage. I have good dumps of 644A,
644B and 654C models. 13,1F,2F options on all.

[james_s]

When I cut the old battery out of the DS1742 in my TDS3000 I was amazed to discover it was only 0.9V and yet it was still holding the data. If I was unable to pull the data off the chip in-situ I'd be tempted to cut the battery out while the board is powered up, or grind into it enough that I could tack wires on.

[nctnico]

The 1486 will never verify twice because the real time clock
shows up as normal ram locations within the chip and are
always changing.
As for corrupting the Dallas DS chips when removing, these
chips on the older scopes have code dates all the way back
to the early 90's and I haven't had one yet that had usable
data after desoldering. Usually just endless strings of 55 55 55
AA AA AA and random garbage. I have good dumps of 644A,
644B and 654C models. 13,1F,2F options on all.

The problem is that desoldered pins don't make good contact in a programmer. I had similar problems and soldering the devices into a turned-pin socket resulted in being able to read the data.

[KG7AMV]

I think the boards are way to far gone!! Sad!!! But spent my budget and 3X's the amount of time allotted under the microscope!! Part Out the known Good stuff!!

[madao]

This is why, i  read always   NVSRAM over tektool, very safe and fast.  Reason: NVSRAM is over 20 years and i have make experience with read over programmer:data corruptions is not rare :-/

NVSRAM-Dump, (not from me)
http://www.ko4bb.com/getsimple/index.php?id=manuals&dir=01_ROM_Images_and_Drivers/Tek_TDS544A