Identify SO-8 chip in Nikon NE-20S Digital Theodolite

[ahakman]

Can anyone identify the "701 B43" chip in this picture?
I'm pretty sure it's a 5v regulator, but I can't find any datasheet that would match the pinout, or any datasheet for a "701" that's a voltage regulator.

The pinout is
1: +9V in (from 6 AA batteries in series)
2: GND
3: to pin 7
4: N/C
5: N/C
6: N/C
7: to pin 3
8: +5 out

On the PCB, pin 3 connects to pin 4 with a trace under the chip, but pin 4 has been cut off the IC, so it's N/C now. It looks like pin 7 originally connected to +5 as well as pin 8, but the pin 7 trace to +5 rail has been cut on the PCB, and pin 3 connected to pin 7 with the bodge wire. Maybe that helps identify a different regulator that was originally meant to go there.

This is from a Nikon NE-20S digital theodolite.

Applying 9V from a bench supply, the chip gets smoking hot, and draws about 500mA. Feeding +5 directly to the 5V rail, the chip sinks 300mA and gets hot. Removing the chip completely and applying 5V to the 5V rail, the board draws less than 10mA, and the display appears to initialize (starts with one line of dark squares, like an uninitialized character LCD usually does, goes blank, then comes back with 2 lines of dark squares), so I think the CPU is actually running, but the contrast on the display is way too dark, so I just get 2 lines of black squares. Looking into what's driving the contrast line on the LCD next.

I'm sure I could just replace that regulator with any 5v low dropout, low quiescent current regulator, but I'm curious what the original "701" is...

Also, if anyone happens to have any manuals (user, or service) for a Nikon NE-20S, I'd appreciate those as well

[amyk]

That's a pretty obscure piece of kit... there's someone selling the manual for it, but this one is for a similar newer model: https://image.tigersupplies.com/Manuals/NIKHRA73100-20141216003558693.pdf

Given the custom Nikon-marked ICs I would not be surprised if that was also a custom part. Otherwise, a scan through Japanese electronics catalogs of the 90s might reveal what it is.

[SiliconWizard]

I'd say it's probably a TOKO TK11701M. Adjustable voltage regulator. Scanned datasheet can be found online.

I think pin 3 and pin 7 shorted sets it at a +5V output indeed (probably pins DR and VS, although I didn't find a datasheet with the pinout, only a block diagram, but it looks like it!)

[fzabkar]

The TK11701M is on page 3-93 (page 352 of PDF):

http://www.bitsavers.org/components/toko/_dataBooks/1993_Toko_Integrated_Circuits_Databook.pdf

[tooki]

Going by the configuration table in that PDF, they originally designed the PCB to configure the IC for 5.25V 5.0V output, and the cut trace and bodge changes it to 5.5V.

So I’d say you hit the nail on the head, SiliconWizard and fzabkar!

Edit: Oops, I'd misread the original post and missed the original pins 3-4 connection. So yeah, originally laid out for 5.0V.

[ahakman]

Damn you guys are good! That's definitely the chip! I'm sure I NEVER would've found the datasheet!

Looks like it was originally configured to 5V (pin 3 connected to pin4, and pin 7 connected to pin 8 ), and they changed it to 5.5V (pin 3 connected to pin 7).

I had a look at the data lines going to the LCD with a scope last night, the CPU is definitely awake and sending stuff to the LCD, but I'm still not sure why it only displays all black squares, and the data looks "odd". They have all the lines connected to the LCD, so they should be using it in 8 bit mode, but on some of the data bits, things didn't look "right" to me on the scope, but I was too tired to hook up a logic analyzer and actually see what's going on.

I seem to remember Sigrock might have a decoder for Hitachi HD44780U protocol, so might be pretty easy to see if the data to the LCD is sane or not (if I had more than an 8 channel USB "logic analyzer" interface to get the 8 data lines AND the 3 control signals...)

They have the LCD contrast voltage tied straight to GND, which I guess should be ok? There was a spot for a variable resistor between the contrast voltage pin and ground (VR1 on the front side), but it's not populated, and there's a 0 ohm jumper (or what's supposed to be a surface mount resistor with no marking on it that's shorted) to ground in parallel with the missing pot ("R" on the back side).

[amyk]

I'd say it's probably a TOKO TK11701M. Adjustable voltage regulator. Scanned datasheet can be found online.

Any hints on how you found that? Did you comb through Japanese databooks or somehow already knew the company?

[fzabkar]

http://www.smdmark.com/en-US/search/code?id=701
http://chip.tomsk.ru/chip/ChipDoc.nsf/vc1?readform&start=1&view=smd&cat=7&count=500
https://www.google.com/search?q=tk11701+filetype%3Apdf+toko

[ahakman]

Definitely bookmarking smdmark.com and chip.tomsk.ru!! Excellent resources!

In the process of putting the grabbers and naming the channels on the trusty 54645D (2 analog + 16ch logic analyzer MSO) to look at the LCD interface.

Too bad my new to me (and sort of new overall, as everything but the back cover was just replaced by keysight under warranty) DSOX2012A only has 8 digital channels... Boooooo

Also, I see the HD44780 decoder is on the "to-do" list at Sigrock - maybe I'll have to write that one and contribute it.

[SiliconWizard]

I'd say it's probably a TOKO TK11701M. Adjustable voltage regulator. Scanned datasheet can be found online.

Any hints on how you found that? Did you comb through Japanese databooks or somehow already knew the company?

As fzabkar said.

But I actually first searched with a couple of search engines directly, including Google. I guess not everyone will quite get the same results because they significantly depend on your history of searches...

[ahakman]

A quick update on this, I looked at the interface to the LCD - the processor sens a few initialization commands to it right at power up, and after that, it never sets the enable line again to ever talk to the LCD - some of the LCD pins seem to be used for something else too - there's a repeating pattern of stuff on some of them, but never anything useful. Haven't had a chance to trace out the connection between the 2 chips.

With the help of the above resources, I was able to identify other chips on the board - the 953A is a cpu reset monitor - a ha! That could be the problem, it's stuck in reset - nope, it works correctly.

Another small part, U8, is a single gate inverter - it inverts the clock between the lower chip connected to the LCD, and the other chip above - that could be the issue... nope, that works correctly too.

I desoldered (with hot air) the top chip completely from the board - it behaves exactly the same with that chip removed.

Not a lot else to check... the resistor networks / dip switches / selector switches have proper continuity maybe? They all go directly to the top chip, which I'm starting to think is an EEPROM, or contains an eeprom with lookup tables so the device can operate in different units (partially from some of the error messages from the manual for the newer Nikon theodolite linked above).  Maybe something to do with the 3 op amps on the back that actually interface to the optical scales? That's about all there is left to check...

Seems like the only way to recover might be to reverse engineer the optical detector circuit, and replace everything with a new uP - If only I knew exactly how it was supposed to work in the first place...

[fzabkar]

https://en.industryarena.com/forum/any-ideas-these-encoders--112240.html

The 2S210 chip appears to be the uC which processes the output from the encoder(s). I suspect that the part number suffix denotes the mask code version. The real manufacturer of the chip is probably one of the Japanese IC makers, eg Mitsubishi, Hitachi, Fujitsu, NEC.


EDIT: This is most likely a red herring, but ...

This web site suggests that the 2S190-143 chip may be an Oki USART (no mask ROM):
http://8moto8.blog.fc2.com/blog-category-13.html

"2S190-143（MSM82C51A-2RS)"

https://www.cryptomuseum.com/crypto/rs/pu104/files/M82C51_datasheet.pdf
http://pdf.datasheetcatalog.com/datasheet/oki/MSM82C51A-2RS.pdf

However, the package is nothing like the 2S190-137. :-(

[ahakman]

Yes, the 2S210-019 is definitely the encoder processor. I traced the all the pins of it. It connects to

• the op-amps which go to the encoders (I think there's one output which goes to both encoders, and 2 inputs from each encoder)
• the selector switches / dip switches (for outputs in different units, or maybe for different encoder types or configurations)
• and to the other chip - looks like an 8 bit bus + 7 other control signals

Both of the selector switches, which were set to D and 6 did not have the correct outputs on the pins. After turning them around and back to D and 6, they did... interesting...

I think I'm going to put the 2S210-019 back on the board, and remove the 2S190 to make it easier to trace all of it's pins.

[ahakman]

So after all that, I decided to connect the encoders back up to the board, moved one of the axes, and boom, up came the display!
I guess it's the decoder processor that feeds data to the other processor that drives the system, not the other way around. No input from the encoders means it just sits there will all black squares. Guess it's fixed!

The horizontal axis doesn't work, but it has a mechanical problem, so once that's fixed, it probably will.