Author Topic: Dallas DS1486 and DS1250 Replacement designs (Read 18815 times)

cuebus · « **on:** March 08, 2020, 08:44:52 pm »

Hi all,

I've made successful replacements of the Dallas DS1486 and 1250 chips that so many of us have been trying to find replacements for.
Based on extensive research on parts and understanding how these chips work, I've come up with the attached board designs.

My DS1486 uses the DS1384G controller chip, which is an internal engineering part Dallas made for their PowerCap modules when they were still in production. It is identical to the DS1384FP as far as I can tell, with the exception of being in a 48 pin LQFP rather than 44 pin MQFP. I determined the pinout through trial and error and help from the DS1558 datasheet, which is a similar controller in the same package.
I'm not sure where suppliers are getting their DS1384G parts, but the parts I've purchased from this supplier are working correctly:
https://specialtypartsinc.com/products/1-pc-ds1384g-dallas-ic-rtc-clk-calendar-par-48-mqfp-1
There is also a chinese supplier on ebay. I have not tested that part.

My DS1250 uses the DS1312S-2 controller chip which is still in production, and the same part Maxim uses in current production DS1250 modules.

These are not only functional replacements but as close to replicas of the originals that I've been able to make using currently available parts. Battery size, count, and components are, aside from footprints and PCB layout, the same as original.

BOM is attached. The SRAM you use is not critical, as long as it is 5V, low standby power (1uA at ambient temps), and the same pinout.

OshPark will make 3 boards for around $6 shipped.

Some tips:
Measure the battery voltage before you mount them. I've had issues with mouser delivering batteries that are DOA and measure less than a volt.
Solder all surface mount devices first. On the fine pitch devices on the 1486, BE SURE THERE IS NO SOLDER SHORT. My first attempt fried both chips on the board because I actually had a solder bridge BEHIND the leads that I didn't easily see. Use plenty of flux.
Next solder the batteries. Be sure there's clearance enough so that the pcb header pins won't be touching the battery tabs.
Do the pin headers last.
On all devices, make sure not to have too much dwell time with your iron. If you have enough flux, the SMD pins will take solder fully in less than a second. I also recommend using the lowest possible temp on the SMD devices.

After you've constructed the board and washed away flux, be sure to let it rest for quite a while before programming to it. You need the SRAM to come back to ambient temperature internally before programming it.

I would love to hear anyone report back how it's working for them after building them.
There are already many other threads on programming them so I won't include anything about that here.

Good luck!

cuebus · « **Reply #1 on:** March 08, 2020, 08:45:52 pm »

james_s · « **Reply #2 on:** March 08, 2020, 09:37:26 pm »

Excellent, thanks! The NVRAMs in my TDS784 are still working just fine however they're around 20 years old now so they're on borrowed time. This is a much cleaner solution than hacking in external batteries as I've been doing with Dallas chips and will be easily maintainable in the future. The key of course being the RTC chips themselves which are not going to be available forever.

Cubdriver · « **Reply #3 on:** March 09, 2020, 01:12:54 am »

Thanks for posting. While I don't need them right now, I'm tagging this for future reference. Nice work.

-Pat

bitseeker · « **Reply #4 on:** March 09, 2020, 04:21:39 am »

Likewise. Good stuff, cuebus!

Rollin Hand · « **Reply #5 on:** March 10, 2020, 03:56:37 pm »

Can someone sell me 4 pc boards?
Thank you.

james_s · « **Reply #6 on:** March 10, 2020, 04:00:33 pm »

JLC PCB will sell you 5 boards for $2.

Rollin Hand · « **Reply #7 on:** March 10, 2020, 04:09:57 pm »

I need them already build ,I am a retired person from the electronic field and my vision dont respond to me,with batteries.Thanks

james_s · « **Reply #8 on:** March 10, 2020, 04:14:37 pm »

There is a seller of similar boards on ebay, I think he charges $100 for a pair which is not unreasonable. This design here was created for those who like to DIY, I doubt anyone is going to be able to build them for you cheaper than those others are.

cuebus · « **Reply #9 on:** March 10, 2020, 10:27:25 pm »

I'm willing to build boards for $50 for each one myself. I did a breakdown of the cost on another thread and it costs about $21 in materials for each chip in small quantity amounts. Add solder, flux, and cleaning materials, and the total cost is probably around $25. It takes about an hour to make both boards.

Only issue is that I am in the middle of a move, so I won't be able to start making these for about another month to month-and-a-half.

I recommend OSHPark for making the PCBs. They use US fabs so lead times are quicker and US environmental regs are in effect.

I have a single DS1486 prototype that I can sell immediately if anyone wants it. It's an older board design, and I had a slight pin misalignment issue with the design so had to bend the pins slightly inward, but it works fine. It's actually the one in the photo. I'll sell it at cost, $25+shipping. I can leave the TDS540C programming or erase the chip before sending. Either way, even if you do have a TDS540C, I believe you will need to redo the SPC constants if you use someone else's data.

Rollin Hand · « **Reply #10 on:** March 11, 2020, 02:25:53 pm »

can you send me an invoice to paypal.To
Rubyg6178

Thanks

CaptDon · « **Reply #11 on:** April 15, 2020, 01:46:37 pm »

I have spent a large amount of time examining my TDS644A memory map to understand the NVRAM
addressing. There are a large number of variants of the 'A' model and the schematics through TekWiki
have been an incredible help!! My TDS644A has only the DS1286 and the 512K DS1650Y. There are many
zero ohm jumpers on the CPU board for memory configuration. The 'A' model is very different than the
'B' model. The 'A' unit is directly related to the TDS544 series and shares much with that line. The floppy
dump tool from FLYTE on this forum is incredible!! The minimalfloppydumper works best on the 644A and
you can use the .bin file with any sensible programmer to program an aftermarket DS1250 replacement
part. DON'T BUY THE FAKE CHINESE CRAP!!!! I have two .bin dumps of my TDS644A if someone may
need a copy. Mine has the GPIB/FLOPPY/MATH options from the factory. There is hardware steering logic
74F27/74F30/74F32 to allow the first 16 bytes of the DS1286 timer to override the first 16 bytes of the
DS1650Y. These are called out as 'bar NVRAM1' and 'bar NVRAM2'. Note that the U1106 and U1107 is
the same socket, the timer fills only the first 28 pins OR the second NVRAM uses all 32 pins.

CaptDon · « **Reply #12 on:** April 22, 2020, 01:16:17 pm »

I bought the newly designed DS1650Y replacement part from the Ebay seller. He normally
sells the DS1250Y/DS1486 set as a kit but allowed me to buy only the DS1250Y to replace
my DS1650Y. I used the minimalfloppydumper from FLYTE to save my TDS644A data (only
512K needed in this scope) and with some indulgence did get the scope working 100%
perfect with no boot errors!!! A word of CAUTION!! As has been noted by others, unsoldering
the old chip will almost certainly destroy any data in it!!! Create your backup bin FIRST!!!
I used the most care possible when unsoldering and removing the old chip. I placed it
carefully into a programmer/reader and read the old data.....TOTALLY SCRAMBLED!!!
I did retrieve a workable TDS644A .bin file and not sure if it may help any of you folks
with the older related 5XX series or not. Cheers!!! Now, on the fix my TDS644B!!

james_s · « **Reply #13 on:** April 23, 2020, 07:08:55 am »

I built a pair using the PCBs posted here, I had a brain fart and installed a chip backwards on both of them but noticed the error and flipped it around before I tried applying power. In the end both worked perfectly, along with the backup and restore scripts and my scope is working perfectly with the replacements. I got the RTC chip from a China seller, I was concerned it might be a fake but it works fine so I think it's the real deal.

cuebus · « **Reply #14 on:** April 29, 2020, 01:24:38 am »

James, great job on assembly, those look more pro than mine!

Glad it's working for you.
I think it's pretty unlikely to find fakes of the LQFP RTC chip unless they are just remarked junk parts of some other chip... The chip is so specific and it's not like it's an op amp where you can take a handful of various low performance op amps and remark them as higher performance chips and the circuit will probably work, especially in audio applications,...and a number of people will probably never know the difference.

I'd still be curious to know though where suppliers even got this chip. It was an internal Dallas engineering part, as confirmed by Maxim, and never sold to the public. I'm guessing at some point there was a surplus auction somewhere and these found their way into various suppliers' hands.

james_s · « **Reply #15 on:** April 30, 2020, 04:57:30 am »

I think if it was going to be a fake, it would be some random useless part that doesn't even attempt to mimic the correct IC. I remember a few years ago one of the "maker" focused companies got burned on a batch of AVR microcontrollers that turned out to be obsolete voltage regulators for PC motherboards or something like that, re-marked as atmega chips.

My guess would be those Dallas ICs are surplus parts from a company that was building modules for Dallas. Wouldn't surprise me if they farmed out the production at some point.

KG7AMV · « **Reply #16 on:** May 15, 2020, 08:00:27 pm »

100 Thumbs Up..

Thanks

Johnny10 · « **Reply #17 on:** May 15, 2020, 08:10:19 pm »

Great Thread!

Tantratron · « **Reply #18 on:** July 04, 2020, 05:24:34 am »

Has anyone in Europe reproduced or build the NVRAM's kit ?

radioguy123 · « **Reply #19 on:** October 26, 2020, 07:16:54 pm »

CUEBUS are you going to sell kits ? How difficult are these boards to build ?

Rasz · « **Reply #20 on:** October 27, 2020, 05:24:37 am »

Quote from: james_s on March 08, 2020, 09:37:26 pm

The key of course being the RTC chips themselves which are not going to be available forever.

But there is nothing wrong with RTC chips to begin with. I find it a little crazy spending $50 on something fixable by replacing $1 battery (with a dremel or disposable soldering iron tip)

james_s · « **Reply #21 on:** October 27, 2020, 06:47:11 am »

Quote from: Rasz on October 27, 2020, 05:24:37 am

But there is nothing wrong with RTC chips to begin with. I find it a little crazy spending $50 on something fixable by replacing $1 battery (with a dremel or disposable soldering iron tip)

Except when there is, I did have one that failed.

I have bodged on replacement batteries lots of times but I liked the idea of having clean, bare, un-potted and easily serviced modules in my scope that don't require any surgery, and $50 is peanuts in the grand scheme of things. I've spent far more building far less useful hobby projects.

whatboy · « **Reply #22 on:** January 04, 2021, 04:17:31 pm »

Thank you very much Mr. cuebus, for your valuable time and by sharing this info!!!

but what chip is the U1 in the DS1250Y board???

whatboy · « **Reply #23 on:** January 04, 2021, 05:29:49 pm »

Just order 5 and 5 boards online, for 20 bucks with shipping!

whatboy · « **Reply #24 on:** January 13, 2021, 01:26:10 pm »

Got the boards from JLCPCB but the holes are too small for the pin headers...


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

Author Topic: Dallas DS1486 and DS1250 Replacement designs (Read 18815 times)

Share me