Dear all,
I will have to replace the dallas chips (DS1486/DS1650) of my TDS684A.
I have ordered the replacement kits (not fake China chip) from a source I have found in EEVBlog.
While I wait, I have dowloaded the contents of the chips to a floppy using a tool that does that.
My question is about the next step.
I intent to read the chips (just in case) and I will have to write one image to them.
I currently have a TL866A and a GQ-4X, but I have seen posts that say that neither programmers will work.
I can’t afford spending $500 or more in another programmer.
Is there any consensus out there if any of these can be used?
Any special recommendation about programming. Which chips should be used as “target device” for the DS1486 and the DS1650?
Thank you

Hey Roger,

I don't know the answer to your question regarding reading and writing the
NVRAM. But I wonder why you feel you need to save and restore the NVRAM
data.

The calibration information for the scope is not stored in NVRAM, it's in
EEPROM on the acquisition board, so you don't need to worry about the
calibration information. AFAIK the only information of value in the NVRAM
is the options bits, and it's well known how to write to them to enable
whatever options you have and/or want to enable.
Also, if you install the new NVRAM in a socket, you should be able to
safely defer the decision on buying a programmer. In the worst case
scenario, you'll still have the old NVRAM, and to program the new one,
you'd only need to pop it out of its socket.

All that being said, I'd love to know what people here are using to read
and write NVRAMs ;).

Siggi

I think I have the GQ-4X and I looked into it programming another dallas part for the TDS320.  What i remember is not all DALLAS parts are directly supported.  Your not going to need a programming voltage and you have to restrict the addresses copied because they won't verify.
Compare the dallas 1486 to the 27128 or 27C128 chips.  See: http://www.unitechelectronics.com/EPROM_data.htm
the 27128 and 27C128 is supported by the GQ-4X(V4),GQ-4x4
The DS1225Y (64K and DS1235Y (256K) are supported.  They are likely closer, because they don't require Vpp.Battery backed up RAM.

So, look at the datasheets and the programming entries and figure out how to not program the clock area.

Where are you located? I can read, archive, and program these chips if you would like.

These chips are supported by BP Micro Programmers including the BP 1410 and BP 1600 and later programmers. They are likely supported by other programmers I have but I do not have access to my 'usual' resources to be able to verify this.

Sometimes, you can find these older programmers at very reasonable prices but you would need a WinXP computer with a parallel port to be able to use them. The later programmers, with a '10', such as the 1610, connect with USB and would have software that would support a later version of Win.

I'm happy to help if I can. I'm in NW FL USA.

Joe

On 19-Sep-20 15:15, Siggi wrote:

(snippage)

All that being said, I'd love to know what people here are using to read
and write NVRAMs ;).

It's not often I have to, for the very reasons you point out, but
pretty much any of the Data I/O 'Unifamily' programmers (Unisite, 2900,
3900, 3980) will easily handle those chips. I usually end up using my
Unisite. Most of the programmers from Advin should work as well.

For the best flexibility with any device programmer, look for one which
uses a design called 'pin driver.'

Keep the peace(es).



--
---
Bruce Lane, ARS KC7GR
http://www.bluefeathertech.com
kyrrin (at) bluefeathertech dot com
"Quando Omni Flunkus Moritati" (Red Green)

Hi,

what if you just try it out? There is no risk of loosing data or damaging anything, I would say.

I did it some years ago in a Sony G90 beamer, as a precautionary measure. I used a not very special USB-powered programmer (Batronix) I purchased on an online auction. Worked flawlessly first time.

cheers
Martin

Hi Joe,
Thank you for your kind offer, but I am located in Brasil.
I think it would be too complicated to ship the Nvrams to you to have them read and programmed.
Regards,
Roger

On Sun, Sep 20, 2020 at 12:07 PM, Martin wrote:

what if you just try it out? There is no risk of loosing data or damaging
anything, I would say.

Sorry for the intrusion. Kick me out of this thread if you want. I posed the same question in my thread "TDS 3000B-family 'scope (TDS 3014B) has problem with LAN interface ".
I have a similar problem: loading a DS1742W in a TDS3000B-series 'scope from a file. My programmer (TL866II plus) doesn't support the DS1742 directly, although it does support several SRAMs. I think the main problem could be the fact that Vcc for the DS1742W is 3.3 V, as opposed to 5 V for the regular DS1742. Does anybody know if I can safely use 5 V in the TL866II plus for loading the -W NVRAM? The datasheet does not distinguish the DS1742 in this respect, giving absolute values of 6 V for Vcc and Vin max as Vcc + 0.6 V.

Raymond

hi Siigi,
Thank you for your reply.
I had understood that the scope will not boot with "blank" NVRAM's.
Is this correct?
If not I can give the reading and copying a try, as Martin suggested, but that would not a critical operation.
Rogers

Roger,

I agree, probably too much hassle.

I looked at the Advin website but the earliest programmer I see that specifically supports these two chips is the SpeedPro 280A. It is USB connected but does not work with Win 7 or later. Needs WinXP or VISTA.

Happy to help if I can.

Good luck.

Joe

Rogerio,

The Xeltek 610P will also program these. I suspect any programmer that can program a 128K battery backed RAM chip, and can be set to start at 0E hex, should be able to program the chip.

You could take a different out-of-the-box approach and attach the chip to an Arduino, wire it as a 128K static RAM chip and write to it skipping the first 14 bytes (0D hex) used by the timekeeping circuit. Might be faster and cheaper than buying a programmer just to burn a flash. Just remember to skip the first 14 bytes as they are used by the timekeeping circuit. 0E hex and higher is RAM. I suspect the data you need to program in is much less than the 128K. You only need to wire up as much RAM address as needed to cover the data you need to program. Pull the rest of the address lines down. Take a look at the data and see how much there is and if there is a serial number or MAC address at the start of the data file. You might just need to add a serial number to get the scope back up.

Regards
Tony

Raymond,

The info in the datasheet is probably not clear because "you should not need to know this". Based on looking at this and other family datasheets it seems as if the difference between the W and non-W version is the power fail voltage level. Both chips seem to be ok with 6V on any pin so I suspect they both work on 5V but the W version should not be used in a 5V environment because of the power fail level being so low. This is just a guess from looking at the datasheets so it is your choice to try it. I would probably pop it into my Xeltek or Unisite and program it that way but if I did not have either, I would try programming as a 5V part.
Alternately, if you are worried, you could try the Arduino driving SRAM method I mentioned to Rogerio on this thread. In that case you could drive it at 3.3V.

I ran into a similar won't work problem with one of my HP 16xx logic analyzers and its LAN interface. Turns out if the MAC address is not programmed in Flash (battery died in a timekeeping device), the LAN interface will not come up at all. I just programmed in a typical value for an HP device (random MAC) and it came right up. Guess this is the same as your problem.

Regards
Tony

On Mon, Sep 21, 2020 at 08:31 AM, tgerbic wrote:

The info in the datasheet is probably not clear because "you should not need
to know this". Based on looking at this and other family datasheets it seems
as if the difference between the W and non-W version is the power fail voltage
level. Both chips seem to be ok with 6V on any pin so I suspect they both work
on 5V but the W version should not be used in a 5V environment because of the
power fail level being so low. This is just a guess from looking at the
datasheets so it is your choice to try it. I would probably pop it into my
Xeltek or Unisite and program it that way but if I did not have either, I
would try programming as a 5V part.
Alternately, if you are worried, you could try the Arduino driving SRAM method
I mentioned to Rogerio on this thread. In that case you could drive it at
3.3V.

I ran into a similar won't work problem with one of my HP 16xx logic analyzers
and its LAN interface. Turns out if the MAC address is not programmed in Flash
(battery died in a timekeeping device), the LAN interface will not come up at
all. I just programmed in a typical value for an HP device (random MAC) and it
came right up. Guess this is the same as your problem.

Hi Tony,
Thanks for your comments. 5 V seemed to be OK for the DS1742W because the data sheet made no distinction between the 3.3 V and 5 V version re. Vcc max. and the idea that the difference in Vpf did not indicate any other differences in the chip.
That was until I saw the data sheet of newer devices, where a Vcc max of 4.6 V is specified for the 3.3 V part. Could all be the reason you mentioned but I have no spare DS1742W in case it dies and can't easily get one, although alternatives exist. That's the main reason for being so careful and hesitant.
I decided to connect/cloak my DS1742W as a DS1750W, although the latter is much larger but it's supported in my programmer (TL866II Plus). Connecting the DS1742W seems quite easy, with most pins straight through and just two wiring changes.
I'll keep your hint re. an Arduino in mind.

ISTR a similar problem with HP's 16702A/B LA's.

Raymond

On Mon, Sep 21, 2020 at 11:14 AM, Raymond Domp Frank wrote:

I decided to connect/cloak my DS1742W as a DS1750W, although the latter is
much larger but it's supported in my programmer (TL866II Plus). Connecting the
DS1742W seems quite easy, with most pins straight through and just two wiring
changes.

Earlier today, I did as I described above and "programmed" the DS1742W successfully at 3.3 V on a TL866II Plus.

Raymond