Post Reply 
HP-48 FRAM Cards
07-16-2016, 02:59 AM
Post: #1
HP-48 FRAM Cards
A fellow member of this forum recently emailed me about FRAM cards sold on EBAY that do not require a battery. (Example)

These cards look to be naked PCBs (no protection whatsoever). As such, wouldn't they be quite vulnerable to damage due to drops or static discharge?

Those of you who own and use FRAM cards for your 48GX, how do you feel about them?

What are the caveats compared with the traditional "battery required" SRAM cards?

Pricing on EBAY also seems odd. From the same seller, I see a "no battery required" 128k FRAM card for $138 while a 2000k FRAM card "requiring a battery" costs $149, and a 512k FRAM card "no battery required", also from the same seller, costs $179. It doesn't make much sense.

Thank you.
Find all posts by this user
Quote this message in a reply
07-16-2016, 04:15 AM
Post: #2
RE: HP-48 FRAM Cards
The cards with batteries are RAM, not FRAM, and $40 is the difference between a 128K FRAM chip and a 512K one.

I don't have FRAM cards for my 48's, but I do have a FRAM module for my 71B. The only caveat is that it can't be erased by removing a battery.
Find all posts by this user
Quote this message in a reply
07-16-2016, 04:21 AM
Post: #3
RE: HP-48 FRAM Cards
(07-16-2016 04:15 AM)Dave Frederickson Wrote:  The only caveat is that it can't be erased by removing a battery.
But is that really a "negative"? :-)

I still wonder about durability/longevity though. The cards ship in an antistatic bag for a reason. Then you remove the card (hopefully while grounded), insert it into your 48GX, and then forget to be grounded and remove the card one day. Seems like a lot of card swaps could potentially destroy one of the chips via static if the user isn't very careful.
Find all posts by this user
Quote this message in a reply
07-16-2016, 07:07 AM
Post: #4
RE: HP-48 FRAM Cards
I have a 128 kb FRAM card in one of my HP48GX without a back-up battery. I haven't experienced any issues with it & purchased it from probably the same Ebay seller you are looking at.

Have noticed that pricing has changed quite a bit over the last couple of years with the larger cards becoming relatively cheaper. Some reasons for this may be that the 128 kb size is more popular as it can be merged with main RAM (as I do) & is probably not much cheaper to manufacture.

Cheers,

Michael
Find all posts by this user
Quote this message in a reply
07-16-2016, 12:20 PM
Post: #5
RE: HP-48 FRAM Cards
(07-16-2016 02:59 AM)JDW Wrote:  These cards look to be naked PCBs (no protection whatsoever). As such, wouldn't they be quite vulnerable to damage due to drops or static discharge?

Presumably once it is installed in your 48, it shouldn't be vulnerable. It seems that you would want to leave it installed most of the time, and as such it will be protected by the calculator itself.
Find all posts by this user
Quote this message in a reply
07-16-2016, 12:24 PM
Post: #6
RE: HP-48 FRAM Cards
Thank you for your replies.

One last question then.

For those of you who use cards in your 48GX, do you have a 128k card merged with base ram and then have a second card that is much larger for storage? Or do you find a single 128k card (merged) to be sufficient for your programs (Meta Kernel, String Writer, and such)?
Find all posts by this user
Quote this message in a reply
07-16-2016, 04:39 PM
Post: #7
RE: HP-48 FRAM Cards
(07-16-2016 04:21 AM)JDW Wrote:  
(07-16-2016 04:15 AM)Dave Frederickson Wrote:  The only caveat is that it can't be erased by removing a battery.
But is that really a "negative"? :-)
In my book, this would be a very big plus, even more so for the larger cards.

I once lost a lot of stuff stored in my (no-name) 2 MB SRAM card, when I didn't use the 48GX for a while (after switching to the 50g). For some odd reasons they designed in a tiny CR1216 cell, although there would have been enough room for a CR2320. #-| The CR1216 needs to be replaced more than once a year - really annoying.

Not much of an issue for my (Epson OEM) 128 KB SRAM card, as its CR2016 cell lasts for many ages.

Regarding FRAM, given that FRAM still has a limited (though high) number of write-erase cycles, I would not consider it perfect as RAM extension (in a device where the built-in RAM is battery powered, anyway), but it certainly sounds great for cards used for data storage.

Greetings,

Matthias


--
"Programs are poems for computers."
Find all posts by this user
Quote this message in a reply
07-16-2016, 09:18 PM
Post: #8
RE: HP-48 FRAM Cards
(07-16-2016 04:39 PM)matthiaspaul Wrote:  
(07-16-2016 04:21 AM)JDW Wrote:  But is that really a "negative"? :-)
In my book, this would be a very big plus, even more so for the larger cards.

I agree, but as so many calc reset techniques include the step, "remove the battery", not being able to do that was the only caveat I could come up with.

One other thing, it looks like the FRAM modules are only available in the smaller capacities.

Dave
Find all posts by this user
Quote this message in a reply
07-17-2016, 09:23 AM
Post: #9
RE: HP-48 FRAM Cards
For those of you who use cards in your 48GX, do you have a 128k card merged with base ram and then have a second card that is much larger for storage? Or do you find a single 128k card (merged) to be sufficient for your programs (Meta Kernel, String Writer, and such)?
Find all posts by this user
Quote this message in a reply
07-23-2016, 12:03 AM
Post: #10
RE: HP-48 FRAM Cards
I have 3 of the 128k fram cards. 2 worked 1 was DOA, seller was good about returning. I've since learned some of the caveats of fram memory here are the biggies:

Pros:
-Non volatile no battery required. I've verified this with my cards and haven't had problems loosing memory when out of the calculator.
-Low power consumption.

Cons:
-Fram memory has a destructive read mechanism, meaning it can only be read a finite amount of times ~(10^14) before the memory is trash.

For the above reason I use mine primarily as battery free backups that aren't used daily. If I loose my ram card info I just pop in the fram card and restore the port. I would be weary of using these for long term heavy reading or writing situations.
Find all posts by this user
Quote this message in a reply
07-23-2016, 12:19 AM
Post: #11
RE: HP-48 FRAM Cards
"READ" destroys FRAM, or did you mean "WRITE"?

Even so, I found this FRAM vs. FLASH article interesting. And here's an article on power consumption. And this YouTube Video shows FRAM vs SRAM vs EEPROM. This TI PDF says FRAM has 1 trillion READ/WRITE cycles.

These articles and videos, when viewed along side your experience, would indicate that FRAM used in 48GX cards is somehow flawed or inferior. Wouldn't you agree?
Find all posts by this user
Quote this message in a reply
07-23-2016, 12:46 AM (This post was last modified: 07-23-2016 12:50 AM by Dave Frederickson.)
Post: #12
RE: HP-48 FRAM Cards
Thanks, Matthias, for those FRAM articles. The first article states that FRAM is good for >10^14 write cycles or >150,000 years of continuous data logging.

For FRAM71 owners, the chip used, a Cypress FM22L16, is good for more than a trillion read/write cycles and data retention is rated at 151 years (Really! Had to eek out that extra year to beat the competition with only 150 years.).

So FRAM's disadvantages are higher cost, lower densities than flash, and storage capacity limitations (according to Wikipedia).

(07-23-2016 12:03 AM)snrowe Wrote:  Cons:
-Fram memory has a destructive read mechanism, meaning it can only be read a finite amount of times ~(10^14) before the memory is trash.

Incorrect! A destructive read mechanism means that the data is lost when read and must be followed by a write operation to restore the data, like magnetic core memory. This has nothing to do with endurance.

Dave
Find all posts by this user
Quote this message in a reply
07-23-2016, 12:51 AM
Post: #13
RE: HP-48 FRAM Cards
Yes READ and WRITE BOTH wear out FRAM. 1 Trillion cycles is on the order of what I read as well, this applies to read and write. The act of reading FRAM erase the data so every read is followed by a right to rewrite the data back.

See article below for some comparison with MRAM
http://www.google.com/url?sa=t&rct=j&q=&...VQumFXPwcQ

I agree this makes FRAM less desirable for the HP48GX or in my case the HP48SX. I only learned these facts after buying the cards, so I now use the for back up purposes and use SRAM for normal use

(07-23-2016 12:19 AM)JDW Wrote:  "READ" destroys FRAM, or did you mean "WRITE"?

Even so, I found this FRAM vs. FLASH article interesting. And here's an article on power consumption. And this YouTube Video shows FRAM vs SRAM vs EEPROM. This TI PDF says FRAM has 1 trillion READ/WRITE cycles.

These articles and videos, when viewed along side your experience, would indicate that FRAM used in 48GX cards is somehow flawed or inferior. Wouldn't you agree?
Find all posts by this user
Quote this message in a reply
07-23-2016, 12:58 AM
Post: #14
RE: HP-48 FRAM Cards
(07-23-2016 12:46 AM)Dave Frederickson Wrote:  Thanks, Matthias...
JDW says, "You're welcome." :-)


(07-23-2016 12:51 AM)snrowe Wrote:  Yes READ and WRITE BOTH wear out FRAM. 1 Trillion cycles is on the order of what I read as well, this applies to read and write.

The only thing that PDF educated me on was temperature. So we shouldn't keep FRAM cards in a room higher than 85°C. That's good to know. As the mercury rises above 80°C, I think I'll switch on the A/C. :-)

Seriously, I don't see the negative you are mentioning. The data shows FRAM to have greater endurance than Flash, and most computers use SSDs now. So in effect, you're safer using FRAM in a 48GX than a computer is using its SSD! How is that a knock against FRAM in a 48GX?
Find all posts by this user
Quote this message in a reply
07-23-2016, 01:00 AM
Post: #15
RE: HP-48 FRAM Cards
Hi David,

I researched FRAM about a year ago and the information I came across claimed FRAM has a destructive read which is followed by a write to write the data back after reading. I don't have all the references handy as I read up on it a while ago, but the article I posted above claims FRAM is destructive read.

If you have information to the contrary I'd love to hear it as I'd like to put my cards back in action without worrying about wearing out the memory from running programs off the card.

(07-23-2016 12:46 AM)Dave Frederickson Wrote:  Thanks, Matthias, for those FRAM articles. The first article states that FRAM is good for >10^14 write cycles or >150,000 years of continuous data logging.

For FRAM71 owners, the chip used, a Cypress FM22L16, is good for more than a trillion read/write cycles and data retention is rated at 151 years (Really! Had to eek out that extra year to beat the competition with only 150 years.).

So FRAM's disadvantages are higher cost, lower densities than flash, and storage capacity limitations (according to Wikipedia).

(07-23-2016 12:03 AM)snrowe Wrote:  Cons:
-Fram memory has a destructive read mechanism, meaning it can only be read a finite amount of times ~(10^14) before the memory is trash.

Incorrect! A destructive read mechanism means that the data is lost when read and must be followed by a write operation to restore the data, like magnetic core memory. This has nothing to do with endurance.

Dave
Find all posts by this user
Quote this message in a reply
07-23-2016, 01:04 AM
Post: #16
RE: HP-48 FRAM Cards
(07-23-2016 12:58 AM)JDW Wrote:  
(07-23-2016 12:46 AM)Dave Frederickson Wrote:  Thanks, Matthias...
JDW says, "You're welcome." :-)


(07-23-2016 12:51 AM)snrowe Wrote:  Yes READ and WRITE BOTH wear out FRAM. 1 Trillion cycles is on the order of what I read as well, this applies to read and write.

The only thing that PDF educated me on was temperature. So we shouldn't keep FRAM cards in a room higher than 85°C. That's good to know. As the mercury rises above 80°C, I think I'll switch on the A/C. :-)

Seriously, I don't see the negative you are mentioning. The data shows FRAM to have greater endurance than Flash, and most computers use SSDs now. So in effect, you're safer using FRAM in a 48GX than a computer is using its SSD! How is that a knock against FRAM in a 48GX?

Quote from article on page 3:
"A read operation in a FRAM is destructive because it requires switching the polarization state in order to sense its state. The read operation has to restore the polarization to its original state after the initial read which adds cycle time to the read operation."

Also page 6 the table states FRAM has a destructive read.
Find all posts by this user
Quote this message in a reply
07-23-2016, 01:22 AM
Post: #17
RE: HP-48 FRAM Cards
(07-23-2016 01:00 AM)snrowe Wrote:  Hi David,

I researched FRAM about a year ago and the information I came across claimed FRAM has a destructive read which is followed by a write to write the data back after reading. I don't have all the references handy as I read up on it a while ago, but the article I posted above claims FRAM is destructive read.

If you have information to the contrary I'd love to hear it as I'd like to put my cards back in action without worrying about wearing out the memory from running programs off the card.

(07-23-2016 12:46 AM)Dave Frederickson Wrote:  Thanks, Matthias, for those FRAM articles. The first article states that FRAM is good for >10^14 write cycles or >150,000 years of continuous data logging.

For FRAM71 owners, the chip used, a Cypress FM22L16, is good for more than a trillion read/write cycles and data retention is rated at 151 years (Really! Had to eek out that extra year to beat the competition with only 150 years.).

So FRAM's disadvantages are higher cost, lower densities than flash, and storage capacity limitations (according to Wikipedia).


Incorrect! A destructive read mechanism means that the data is lost when read and must be followed by a write operation to restore the data, like magnetic core memory. This has nothing to do with endurance.

Dave


Snrowe,

Okay, so a write-after-read architecture does have more read and write cycles compared to flash, but FRAM can endure more than 10 billion times the number of read/write cycles as flash. 150,000 years is a lot longer than your 48SX will ever last.

Dave
Find all posts by this user
Quote this message in a reply
07-23-2016, 01:35 AM
Post: #18
RE: HP-48 FRAM Cards
(07-23-2016 01:22 AM)Dave Frederickson Wrote:  
(07-23-2016 01:00 AM)snrowe Wrote:  Hi David,

I researched FRAM about a year ago and the information I came across claimed FRAM has a destructive read which is followed by a write to write the data back after reading. I don't have all the references handy as I read up on it a while ago, but the article I posted above claims FRAM is destructive read.

If you have information to the contrary I'd love to hear it as I'd like to put my cards back in action without worrying about wearing out the memory from running programs off the card.


Snrowe,

Okay, so a write-after-read architecture does have more read and write cycles compared to flash, but FRAM can endure more than 10 billion times the number of read/write cycles as flash. 150,000 years is a lot longer than your 48SX will ever last.

Dave

Flash doesn't have a destructive read, so reading Flash memory doesn't wear it out, only writing to it can do that. 150 years is the data retention time. Fatigue life is driven by number of read and writes while on Flash its just driven by the number of writes.
Find all posts by this user
Quote this message in a reply
07-23-2016, 01:40 AM
Post: #19
RE: HP-48 FRAM Cards
(07-23-2016 01:04 AM)snrowe Wrote:  
(07-23-2016 12:58 AM)JDW Wrote:  JDW says, "You're welcome." :-)



The only thing that PDF educated me on was temperature. So we shouldn't keep FRAM cards in a room higher than 85°C. That's good to know. As the mercury rises above 80°C, I think I'll switch on the A/C. :-)

Seriously, I don't see the negative you are mentioning. The data shows FRAM to have greater endurance than Flash, and most computers use SSDs now. So in effect, you're safer using FRAM in a 48GX than a computer is using its SSD! How is that a knock against FRAM in a 48GX?

Quote from article on page 3:
"A read operation in a FRAM is destructive because it requires switching the polarization state in order to sense its state. The read operation has to restore the polarization to its original state after the initial read which adds cycle time to the read operation."

Also page 6 the table states FRAM has a destructive read.

I see, in this case, "destructive read" has two different meanings. Magnetic core memory and FRAM employ a destructive read architecture where it's necessary to change the state of a bit in order to read the value. The original value then needed to be rewritten back. This is destructive in that the data is lost when read, but does not wear out the device. Core memory had no such characteristic. With respect to endurance, reading and writing reduces the lifetime of FRAM and flash, so one could say that FRAM is more destructive than flash because of the write-after-read architecture, but FRAM still has much better endurance. Does that help?

Just to clarify things a little more, it's not necessary to change the state of all the bits in order to read them. With core memory, and forgive me if I get the polarity incorrect as it's been a while, to read the data all ones are written to a register. A sense line then detects a pulse from the bits which were zeros and changed state. Then the zeros are written back.

Dave
Find all posts by this user
Quote this message in a reply
07-23-2016, 02:01 AM
Post: #20
RE: HP-48 FRAM Cards
(07-23-2016 01:22 AM)Dave Frederickson Wrote:  Okay, so a write-after-read architecture does have more read and write cycles compared to flash, but FRAM can endure more than 10 billion times the number of read/write cycles as flash. 150,000 years is a lot longer than your 48SX will ever last.

Ditto what Dave said, which is what I've been trying to say all along.

Now if we could just fine a FRAM card seller who would sell them at reasonably $50 prices! :-) It's not like we're asking for 128GB here. It's 128kb!!!
Find all posts by this user
Quote this message in a reply
Post Reply 




User(s) browsing this thread: 1 Guest(s)