EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Galen on March 12, 2019, 04:40:08 am
-
Have replaced the NVRAM with new parts for my TDS scope. Now these parts are still availabe on internet. Maybe some day, can't find them anywhere. So I decided to open the original NVRAM, and find out If I can attach a battery socket on it.
Finally,using a hand saw, opened it and dismantled the battery.
-
Two batteries inside, but the positive lead is not connected each other. So Have to use 2 sockets, with negative lead connect together. Solder the wire and glue it.
-
Put them back to my TDS scope, using floppy to write back the image, start the machine, it works! So I Have the Backup now.
-
It looks ugly because of the glue, but works well. No need to buy the NVRAM any more. When the battery die, just replace the battery, write back the image, machine recovered.
-
good work i also thinking the same, cutting it open in half using hand saw or rotating drill or something, but one day will... here's the original video i first saw its done on tektronix TLA704 DS12887 NVRAM, but not fully open it, just minimal invasion...
https://www.youtube.com/watch?v=daoxD-SS3ZU (https://www.youtube.com/watch?v=daoxD-SS3ZU)
-
Thanks mechatrommer for the video. looks it is more easy using iron to dig to the leads. I will try this way next time. The leads location is at the 4 corner for DS1486 and DS1650Y.
-
he must already broke one DS12887 before to know which pin goes where. we can do his technique if every D**** NVRAM have same battery layout, what if not? thats why i prefer your method, open all and sand bit by bit until somethings revealed.
-
I knew that there was 2 batteries in them.
https://www.eevblog.com/forum/repair/tds-684c-nvram-backup/msg1177480/#msg1177480 (https://www.eevblog.com/forum/repair/tds-684c-nvram-backup/msg1177480/#msg1177480)
What I don't know is if they use a couple of schottky diodes as sort of a load balancing scheme, or if the chip uses one battery until it reaches a certain low voltage, then switches over to the other one.
Did you happen to measure the battery voltages before tossing them?
They seem to last FAR longer than anyone expected:
https://www.eevblog.com/forum/repair/tektronix-tds-684c-won_t-boot-front-panel-all-leds-lit-up-only/msg1277714/#msg1277714 (https://www.eevblog.com/forum/repair/tektronix-tds-684c-won_t-boot-front-panel-all-leds-lit-up-only/msg1277714/#msg1277714)
On this one, it seems like it was still working on the first battery which was 2.97V as the other one was 3.28V, suggesting a switch-over scheme.
-
good work i also thinking the same, cutting it open in half using hand saw or rotating drill or something, but one day will... here's the original video i first saw its done on tektronix TLA704 DS12887 NVRAM, but not fully open it, just minimal invasion...
I got to wonder why anyone would bother - you can still buy those chips for $10 or so...
-
I knew that there was 2 batteries in them.
https://www.eevblog.com/forum/repair/tds-684c-nvram-backup/msg1177480/#msg1177480 (https://www.eevblog.com/forum/repair/tds-684c-nvram-backup/msg1177480/#msg1177480)
What I don't know is if they use a couple of schottky diodes as sort of a load balancing scheme, or if the chip uses one battery until it reaches a certain low voltage, then switches over to the other one.
Did you happen to measure the battery voltages before tossing them?
They seem to last FAR longer than anyone expected:
https://www.eevblog.com/forum/repair/tektronix-tds-684c-won_t-boot-front-panel-all-leds-lit-up-only/msg1277714/#msg1277714 (https://www.eevblog.com/forum/repair/tektronix-tds-684c-won_t-boot-front-panel-all-leds-lit-up-only/msg1277714/#msg1277714)
On this one, it seems like it was still working on the first battery which was 2.97V as the other one was 3.28V, suggesting a switch-over scheme.
Yes, measured the 2 battery, 1 is around 2.0V, another is still around 3.0V. So very possible a battery switching circuit is on the NVRAM.
-
good work i also thinking the same, cutting it open in half using hand saw or rotating drill or something, but one day will... here's the original video i first saw its done on tektronix TLA704 DS12887 NVRAM, but not fully open it, just minimal invasion...
I got to wonder why anyone would bother - you can still buy those chips for $10 or so...
i just ordered few of these refurbished (not new) dallas nvram from ebay, yeah its about $10 each for a pair. iirc in digikey its 2-5X of that cost, and then there's
1) 1-2 weeks waiting time. changing battery is $1 easy available local store in 1 day.
2) why waste perfectly working ram when the faulty is only dead battery?
3) who knows 10-20 years from now, there is no dallas nvram supply anymore?
4) some equipment relies on the ram content from factory (calibration data) to work, such as advantest SA. changing nvram unit requires uploading the content everytime, its lucky if nvram programmer still around, putting battery to outside, we can switch battery before dead in some parallel wiring way to avoid losing power (erased content).
ymmv.
-
I got to wonder why anyone would bother - you can still buy those chips for $10 or so...
For how long and have you checked the date ?
My last Dallas from Ebay buy last year gave me a part from 2006, it was good but the batteries are already 12 year old. :(
If there is a way , anyway to prevent a dependency on NOS parts with ticking left time failures inside them I would grab it ;)
-
I got to wonder why anyone would bother - you can still buy those chips for $10 or so...
For how long and have you checked the date ?
My last Dallas from Ebay buy last year gave me a part from 2006, it was good but the batteries are already 12 year old. :(
If there is a way , anyway to prevent a dependency on NOS parts with ticking left time failures inside them I would grab it ;)
The DS12887+ is still an active part. Why would you buy crap from Ebay?
-
I got fed up with these damn Dallas NVRAMs and developed a FRAM replacement for it.
Cypress makes FRAM with the same async parallel bus and 5V operating voltage. The timings are close enough to Dallas NVRAM while providing 100 year data retention. Sometimes there are even pin compatible PDIP versions of them but i couldn't find them for my case so i created a SMD to PDIP adapter board that fits a larger than necessary FRAM chip onto all the Dallas footprints i needed. Tho obviously for the RTC ones the clock will stand still, hopefully the self test does not detect that (The ones i replaced didn't have a clock).
So far running these in a HP 3458A and it works like a charm.
-
The DS12887+ is still an active part. Why would you buy crap from Ebay?
Then you are lucky (for now) but my Tekscope had older EOL dallas parts.
-
i rechecked DS1230AB/Y in digikey (the one i need) its 5X the cost compared to ebay, DS1250AB? its 23X. i guess the reason is we love lucky draw, with hope that we are lucky. ebay is a better place for this compared to our annual dinner event :palm: in case its broken, i'm going to hack it anyway and $4/chip is not much hurt, i can buy from digikey again if lesson learnt. the seller is honest enough to claimed them as "Renovation" chips. ymmv.
-
i wouldnt buy these "new",
last time i got one from RS it was 9years old!
i designed a circuit on stripboard a few years back that not only uses fram, but also retains the old module and uses some ttl to bank the clock registers in over the fram.
it was an ugly mess, but it worked.
maybe i will do a fresh update sometime with smd parts and a gal instead of 2 ttl's
-
I know old post but anyone ever de-package these and take precision measurements
MY CNC Mill is itching to machine some precision holes? After I clean it off from reorganization of the shop.
(http://www.stevenrhine.com/wp-content/uploads/2020/05/20200512_150408-1024x768.jpg)
-
Well with a sweet machine like that i suppose you could just take a really big end mill and just slowly mill away the whole top of the chip in tiny layers until you see some shiny metal.
-
Berni, that is the plan.. I would like to poke some precision holes and solder.. But i did get a good read of the DS1650Y
(http://www.stevenrhine.com/wp-content/uploads/2020/05/544A-DS1650Y-Verify-3X-LV-1024x576.jpg)
File http://www.stevenrhine.com/wp-content/uploads/2020/05/TDS-544A-DS1650Y-NVRAM-Verify-3X.zip (http://www.stevenrhine.com/wp-content/uploads/2020/05/TDS-544A-DS1650Y-NVRAM-Verify-3X.zip)
-
I am investigating the quantity of battery inside Dallas NVSRAM.
Please let me know if what I'm investigating is wrong or if you have anything to add
DS1245 32pin, 2 batteries
DS1486 32pin, 2 batteries
DS1650Y 32pin, 2 batteries
DS1230Y 28pin, 2 batteries
DS1250Y 32pin, 1 Battery
DS1286 28pin, 1 Battery
DS1644 28pin, (?) Battery
DS1630AB 28pin, (?) Battery
I will replace these batteries first, and if it fails, I will have to purchase more (at least one).
The 1230Y has already failed.
The 1230Y was removed from the board as shown in the attached figure, and six legs were cut, and it seems that the internal circuit was damaged due to excessive surgery while performing a leg suture. (A circuit passes between legs 2 & 3.)
Of course, data was obtained before and after the operation using TL866plus2, but start is slightly different when comparing the data.
There is a difference between 01111111 & 00101001.
The above difference means point & parenthesis.
Do you have x-ray photos of DS1644 & 1630AB?
Or let me know how many batteries you have.
-
https://www.antiqueradios.com/forums/viewtopic.php?f=8&t=352206 (https://www.antiqueradios.com/forums/viewtopic.php?f=8&t=352206)
I believe in the x-ray image of the DS1230Y-120 I linked, and I know that there will be two batteries, and I've drilled it.
The picture below is the result.
However, there is only one battery.
I worked up and down the same height as the x-ray image, but no bottom.
What I've been working on is the DS1230Y-150 +, is it completely different from the one I linked above?
The battery side shown for reference is GND (-).
It is pin14 and short at the bottom right.
Now we need to find the Battery plus.
Drilling is done with too much powder, so now you have to cut it with a knife.
Does anyone know where Battery Plus is?
-
All the button and coin cells that I've seen have a steel case. A small rare-earth magnet should show you where they are.
Also, on all the button and coin cells that I've seen, the contact that appears on the edge of the battery is the positive contact. The contact that is just a round disk is the negative side. In your picture, if the round disk is the negative contact, you should be able to find the positive side by digging just outside of the disk.
I've been able to measure the battery voltage on a few of these devices by using a small magnet to find the edge of the battery and then drill in from the side with a hand-driven drill until I got to the positive contact. The negative side of the battery was connected to the ground pin of the device.
-
edpalmer42, thanks for the answer.
I've watched a lot of videos on the internet all day looking for Positive(+).
There were two ways.
The first method is to extract the battery by digging the visible negative(-) (this method will take a really long time).
The second method is to cut out the battery between pin1 & pin28 like DS1230Y-150+.
I decided to do both.
I worked hard for 2 hours.
Finally, a positive(+) bottom was seen and the voltage was measured. DC 3V. It is normal.
Collapse
But it's okay. There is no regret because I saw what I was curious about.
The DS1230Y-150+ is a part included in the “Unicorn UDL-200L” DC Electronic Load.
Now we have to look at the circuit diagram again and worry again.
-
Hi I have a tds644a I think the battery in the nvram is about to fail, how did you copy the image to floppy disk and how did you copy it back to the chip. Please can you help
-
:-+ :-+ :-+