Author Topic: Yes, another 3458A repair and NVRAM replacement to RAMTRON FRAM (Read 25073 times)

quarks · « **Reply #25 on:** January 16, 2017, 06:37:57 pm »

my programmer shows this directly after programming.
Data on every other address is written correctly.

bingo600 · « **Reply #26 on:** January 16, 2017, 06:54:04 pm »

A "trick" could be to make the upper half of the FRAM "active" , and program that one w. the Cal data.
That would not solve the mystery of location 0 , but might make it usable.

Ohh... And i suppose you forgot to attach the pic' of your batronix output

/Bingo

Tektron · « **Reply #27 on:** January 17, 2017, 09:51:44 am »

Maybe programmer does it.
Perhaps you insert in programmer fram with modified address bus? Then, possibly, programmer does not know "short" fram chip. And he wrote all + 1byte in address 800h (I don't know why). This additional byte rewrites address 00 with incorrect info because address bus was modified.
For solving may be better will programming full size image. It may be 4 times repeated in ONE file image of ds1220, or one image of ds1220 + 3 "virtual" images (00 or FF filled) in one file. Hard to explain in english, sorry

.

BravoV · « **Reply #28 on:** January 17, 2017, 11:07:16 am »

Quote from: Tektron on January 17, 2017, 09:51:44 am

Hard to explain in english, sorry .

Just type here in Russian, and we will wait Tin for translation.

quarks · « **Reply #29 on:** January 17, 2017, 12:17:29 pm »

I think I found the problem, it must be the slow "Power-Up to First Access (tPU)" of 10ms from the FM1608B/FM16W08.
Older FM1608 were much faster (only 1µS).
Therefore my programmer BX48 is most likely just to fast, so the data in the very first address 00h is not written.
I hope I find a way to solve this without waiting for an update from Batronix.

I reported my finding to Batronix and will give you an update as soon as I have a feedback

quarks · « **Reply #30 on:** January 17, 2017, 03:49:11 pm »

for the documentation, here is the report from BX48 programmer after writing the DS1220 data to FM16W08

TiN · « **Reply #31 on:** January 17, 2017, 05:35:30 pm »

Hmm, where this come from, I wonder? Time to go deeper into the hole, to get all the rabbits out.

quarks · « **Reply #32 on:** January 17, 2017, 06:32:18 pm »

Hello TiN,

what a coincidence, I just tested Todds board with 16W08. DS1220 data was written incl. address 0 and it works in the 3458A without ERRSTR.

I am sure that the "normal" SOIC-DIP28 board will also work, as soon as data can be written to address 0.

bingo600 · « **Reply #33 on:** January 22, 2017, 06:25:44 pm »

@quarks

Did you get your programmer fixed ??

Quote from: TiN on January 17, 2017, 05:35:30 pm

Hmm, where this come from, I wonder? Time to go deeper into the hole, to get all the rabbits out.

@TiN
Where did you get those boards ?

/Bingo

TiN · « **Reply #34 on:** January 23, 2017, 04:06:59 am »

Emm, I got them from Datron 1281 article

quarks · « **Reply #35 on:** January 23, 2017, 08:03:33 am »

Quote from: bingo600 on January 22, 2017, 06:25:44 pm

@quarks

Did you get your programmer fixed ??

I have no response/fix from Batronix so far

bingo600 · « **Reply #36 on:** January 23, 2017, 08:00:37 pm »

Quote from: TiN on January 23, 2017, 04:06:59 am

Emm, I got them from Datron 1281 article

Not very helpfull info , unless one gets sent 2 free boards for reading the article

Any hints , for getting the boards

/Bingo

mimmus78 · « **Reply #37 on:** January 23, 2017, 08:18:46 pm »

Read the article ... where I can signup for the boards?

ManateeMafia · « **Reply #38 on:** January 23, 2017, 10:34:40 pm »

I had the boards made at ITEAD Studio. Attached are the Gerbers.

I think that the FRAM board may go through another rev based on some ideas listed in this thread. But if you must have them now, you can get a batch made.

mimmus78 · « **Reply #39 on:** January 23, 2017, 10:38:53 pm »

My ram chips were just swapped ... I guess I have some time before next pit stop.

Inviato dal mio Nexus 6P utilizzando Tapatalk

nidlaX · « **Reply #40 on:** January 24, 2017, 07:26:26 am »

Thanks for the article! I'm going to grab some old RAMTRON FM1608 in DIP28 from eBay to replace the SRAM in my Keithley 193.

bingo600 · « **Reply #41 on:** January 24, 2017, 03:36:53 pm »

Quote from: ManateeMafia on January 23, 2017, 10:34:40 pm

I had the boards made at ITEAD Studio. Attached are the Gerbers.

I think that the FRAM board may go through another rev based on some ideas listed in this thread. But if you must have them now, you can get a batch made.

@MM
Any hints to the function of new PCB's ?

/Bingo

ManateeMafia · « **Reply #42 on:** January 25, 2017, 01:43:02 am »

I wanted the board to be used as a replacement for three different types of memory.
1. Dual CE Ram JMP1,JMP2-> pos 2. R1,R2 are installed as pullups. R3 not installed. IC2 installed.
2. DS1225 JMP1->pos1 JMP2->nc R1,R2 are installed. R3 not installed. IC2 not installed.
3. DS1230 JMP1,JMP2->pos 1. R1,R2 are not installed. R3 is installed. IC2 not installed.

JMP1 and JMP2 are zero ohm. I used 4.7k on R1,R2. R3 is zero ohm. I should have labeled it a jumper. IC2 is a 74LS00.
R1 and R2 tie the unused address lines high. An alternative design could use a two way switch between VCC and gnd for A13 and A14 for multiple backups.
JMP3 was originally added for in-circuit reading but I decided to just use a wire. I have the additional board for my EPROM programmer that seems to work fine.

The FM1808 can be substituted for cheaper lower capacity FRAM with the additional savings of R1&R2. Any future boards for my Datron products will use the cheaper FRAM.

bingo600 · « **Reply #43 on:** January 25, 2017, 05:36:41 pm »

Quote from: ManateeMafia on January 25, 2017, 01:43:02 am

I wanted the board to be used as a replacement for three different types of memory.
1. Dual CE Ram JMP1,JMP2-> pos 2. R1,R2 are installed as pullups. R3 not installed. IC2 installed.
2. DS1225 JMP1->pos1 JMP2->nc R1,R2 are installed. R3 not installed. IC2 not installed.
3. DS1230 JMP1,JMP2->pos 1. R1,R2 are not installed. R3 is installed. IC2 not installed.

JMP1 and JMP2 are zero ohm. I used 4.7k on R1,R2. R3 is zero ohm. I should have labeled it a jumper. IC2 is a 74LS00.
R1 and R2 tie the unused address lines high. An alternative design could use a two way switch between VCC and gnd for A13 and A14 for multiple backups.
JMP3 was originally added for in-circuit reading but I decided to just use a wire. I have the additional board for my EPROM programmer that seems to work fine.

The FM1808 can be substituted for cheaper lower capacity FRAM with the additional savings of R1&R2. Any future boards for my Datron products will use the cheaper FRAM.

Yes please .....

Can i have 2..3 boards tomorrow ??

/Bingo

rastro · « **Reply #44 on:** January 26, 2017, 12:10:22 am »

The F-RAM memories are marketed as 3.3VDC devices although the datasheets indicate a max voltage of 5.5VDC. Is there any concerns about running these chips on high side of their voltage range? Also does the logic levels follow the power supply level?

I would like to adapt one of these chips for the HP 8753's which have a very poor ram back that typically can't survive more than day or two powered down.

-rastro

nidlaX · « **Reply #45 on:** January 26, 2017, 12:59:22 am »

Quote from: rastro on January 26, 2017, 12:10:22 am

The F-RAM memories are marketed as 3.3VDC devices although the datasheets indicate a max voltage of 5.5VDC. Is there any concerns about running these chips on high side of their voltage range? Also does the logic levels follow the power supply level?

I would like to adapt one of these chips for the HP 8753's which have a very poor ram back that typically can't survive more than day or two powered down.

-rastro

I think the RAMTRON / Cypress design has simply been upgraded over the years from 4.5V - 5.5V to 2.5V - 5.5V with the W model, so there should be no concerns since none are stated in the datasheet.

james_s · « **Reply #46 on:** January 26, 2017, 05:50:57 am »

While you're not supposed to, it's possible to replace the battery in the DS1220 and similar chips. It's a lithium coin cell potted in the block, it's usually up near the top and can be found with a bit of cutting with a dremel. Once you extract it, solder a CR2032 holder to the remains of the pins that went to it.

bingo600 · « **Reply #47 on:** February 01, 2017, 04:06:12 pm »

I guess it's my time now, to install FRAM's

https://www.eevblog.com/forum/repair/3458-ram-test-1-low/

/Bingo

quarks · « **Reply #48 on:** February 12, 2017, 03:35:00 pm »

just a short update.

As already stated, I contacted Batronix about my finding and they were very helpful and responsive.

After several not working updates they asked for a chip and I send a FM16W08.
This weekend I was able to test their solution and it works now.
Unfortunately it is not yet public, because it needs a new Firmware which is not yet officially released.
I will give an update as soon as the fix is publicly available.

PS: I have a few of the ManateeMafia "Dual CE EPROM Programmer Adapter" boards, which I do not need with my programmer. If anyone is looking for these, feel free to send a PM.

rastro · « **Reply #49 on:** February 22, 2017, 11:23:52 pm »

What is everyone doing to make pins for the SOIC-DIP adapter board?


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

Author Topic: Yes, another 3458A repair and NVRAM replacement to RAMTRON FRAM (Read 25073 times)

Share me