32k RAMPAK

Showcase your Organiser hardware developments here
amenjet
Posts: 237
Joined: Tue Jan 03, 2023 7:54 pm

Re: 32k RAMPAK

Post by amenjet »

Having spent some time this morning and failed at trying to get my 256K RAM pack to fit in to a datapack case with a battery, I think it comes down to mechanical constraints. The height is much reduced if there's no battery holder or PCB in there.

Andrew
MartinP
Posts: 52
Joined: Wed Jan 04, 2023 7:51 pm

Re: 32k RAMPAK

Post by MartinP »

Are you using the more compact side entry connectors like these:
https://www.e-tec.com/v5/products/pcb-i ... ry_76.html
These connectors seem difficult to find, but this company has a sales office in Tunbridge Wells. They might have a minimum order quantity, 'though?
Martin P.
User avatar
Martin
Global Admin
Posts: 250
Joined: Mon Jan 02, 2023 5:18 pm

Rampack cases

Post by Martin »

Andrew...

I know what you mean about it all being tight with the large capacity datapack.. Olivier recommended an adjustment to the 'back' of the pack case... I didn't bother making the alterations, I've simply left it off. You can still slide the pack home without the 'back' in place..
.
DSCI0748.JPG
.
Sincerely Martin

PS Zip contains more images of the 'backless' rampack
256K dataack case.zip
You do not have the required permissions to view the files attached to this post.
thesourcerer
Posts: 72
Joined: Sat Jan 07, 2023 1:00 pm

Re: 32k RAMPAK

Post by thesourcerer »

I did manage to get the backs to fit by extending the slots in the sidewalls of the back. Even the original 32k Rampacks (I think) had enlarged slots due to the diameter of the battery.
The problem with Olivier’s was exacerbated by the plastic battery cover, which isn’t necessary.
amenjet
Posts: 237
Joined: Tue Jan 03, 2023 7:54 pm

Re: 32k RAMPAK

Post by amenjet »

MartinP wrote: Sun Feb 12, 2023 12:09 pm Are you using the more compact side entry connectors like these:
https://www.e-tec.com/v5/products/pcb-i ... ry_76.html
These connectors seem difficult to find, but this company has a sales office in Tunbridge Wells. They might have a minimum order quantity, 'though?
Martin P.
Hmm, no, I'm using a standard right angle connecto. It requires some modification of the datapack case. the problem isn't really at that end, it's in the main space of the datapack where the battery has to somehow fit on top of the RAM chip itself. I am unsure what the solution is at the moment. maybe a 2016 battery instead?

Andrew
amenjet
Posts: 237
Joined: Tue Jan 03, 2023 7:54 pm

Re: Rampack cases

Post by amenjet »

Martin wrote: Sun Feb 12, 2023 12:14 pm Andrew...

I know what you mean about it all being tight with the large capacity datapack.. Olivier recommended an adjustment to the 'back' of the pack case... I didn't bother making the alterations, I've simply left it off. You can still slide the pack home without the 'back' in place..

Sincerely Martin

PS Zip contains more images of the 'backless' rampack256K dataack case.zip
I might try that. i actually have two flashpacks in my organiser and I have no case at all on those. They have worked fine for quite a while now. the reason for that is that I don't have many cases, rather than any mechanical issue. They do fit easily in modified cases.

Andrew
Nick Kostelidis
Posts: 12
Joined: Wed Jan 04, 2023 4:47 pm
Location: Athens Greece
Contact:

Re: 32k RAMPAK

Post by Nick Kostelidis »

I have been using rampaks since 1992.
I use them to store data from surveying instruments. I use them because the data write speed is fast. The speed of writing data to data packs and flash data packs is such that it makes data writing slow, sluggish and problematic.
I usually use 64k ram packs.
I can store up to 650 records (measurments) in a 32k ram data pack and up to 1250 records in 64k.
The best for me is to use the 64k ram data pack.
I have worked with both 124k and 256k ram pages.
The problem with these packages is that the more recordings stored in them, the slower the recording becomes. That's why I prefer the 32 and 64 k ram paks.
In addition, it is very quick and easy to delete data from these packages, which is not the case with data packages.
If someone uses them correctly I believe they are safe.
I have been using them for over 30 years.
I've only lost data once when I dropped the Psion, the battery got disconnected and I made the mistake of removing the rampack while the organizer was running.
When one uses it properly it is absolutely SAFE.
I change the batteries in the rampacks every 5 years.
The last two times I've changed batteries in rampaks I've used cr1632 batteries instead of cr2032 batteries. Obviously they have less capacity but are smaller in size. They are easier to install and require absolutely no modification to the original cases. It's just that instead of 5 years I will change them every 3 years.
You do not have the required permissions to view the files attached to this post.
Last edited by Nick Kostelidis on Tue Feb 14, 2023 5:52 pm, edited 5 times in total.
User avatar
Martin
Global Admin
Posts: 250
Joined: Mon Jan 02, 2023 5:18 pm

Flashpack Test Report

Post by Martin »

Nick (and others watching)

Try editing your post above and adding your picture now... I've spotted an error and changed the attachment settngs. - PM me if you are still having difficulties.

Read & Write times to devices in B: and C:
I did a Flashpack testing report (here) for Andrew when he was developing his 512K Flashpack. You will see that the time taken to write data to a flash pack increases (by about 28%) the deeper the file is in/on the pack. I didn't do an in-depth test on RAM but you will see a comparison of about 3% in time to write data. The biggest time related issue with large capacity ram packs is 're-saving' or 'updating' a file where, because the data is stored 'sequentially', the original file is deleted and the other files on the pack are 'shunted up' to file the space created by deleting the file before the 'updated' file is tagged on the end.

Finally.. I didn't know there are any 64K Rampaks... where did you find them? Are the button cell batteries 'centre' pins?

Always in good faith
Martin
Nick Kostelidis
Posts: 12
Joined: Wed Jan 04, 2023 4:47 pm
Location: Athens Greece
Contact:

Re: 32k RAMPAK

Post by Nick Kostelidis »

Martin, take a look at this page.
https://www.jaapsch.net/psion/galpacks.htm#rampak
Yes, there are 64k rampaks.
I have had them for more than 20 years.
Two of them had been given to me by a good friend of mine from Serbia.
I have two more that I don't remember where I got them from.
I have sent to your mail a photo of a CR1632 battery.
https://www.aliexpress.com/item/4001159867457.html
As far as 512 flash data packs are concerned, I use them only to store apps.
I prefer them over data paks.I have 5 pieces of them.
4 of them came from Olivier and one from Andrew.
All of them work fine.
The only difference for me is that Andrew's flash pak requires a case modification.
Lostgallifreyan
Posts: 83
Joined: Thu Jan 12, 2023 8:25 pm

Re: 32k RAMPAK

Post by Lostgallifreyan »

Nick Kostelidis wrote: Tue Feb 14, 2023 8:01 am I change the batteries in the rampacks every 5 years.
The last two times I've changed batteries in rampaks I've used cr1632 batteries instead of cr2032 batteries. Obviously they have less capacity but are smaller in size. They are easier to install and require absolutely no modification to the original cases. It's just that instead of 5 years I will change them every 3 years.
This is good practise. The risk mainly arises from neglect of main battery, causing more drain on the backup to the point where handling the RAM pack can cause data loss. It's not a weakness of the pack design, it just needs care with cell charge state..

It's worth writing the install date on a new cell, I do that, always, so if in doubt I can open up a RAM pack to check how old the cell is. I use a RAM pack in any Organiser I use, and if I could get a bunch of 64K RAM packs I would. I'm not sure why addressing larger ones might be slower, but I'll guess it is because the native integer is 16 bit, so can address up to 64K with no need for extra CPU time for each access. (With Flash, the 'slower' address speed doesn't matter because that's not what slows Flash most, except maybe when reading data.)

Using a deeper cell with less width may be better to maximise volume. Logically that's false because of the ratio with increasing diameter, but there are fixed overheads in what is already a very flat cell! It might run into design constraints to try smaller but deeper cells in new layouts, but it's worth considering.

About the pack lids, I thought they all had the same form, a bit cut out of the wall on each side for a coin cell, even when the lid was for a different type of pack. Easier to make them all the same way... When I replace a coin cell I solder bits of wire to new cells. It's cheap but it takes a bit of care to do it well while minimising heat transfer. Very sharp timing and a good resin flux.. For an insulator I use the 'blister pack' plastic from the card the cells arrive on, and cut with scissors to make it fit. It even comes with a free hole on the middle. :) It's a bit tedious but once in ten years, per pack, usually works. It certainly minimises 'resizing' events. Messing with experimental loads on the top slot is the biggest risk of those anyway.
Post Reply