Advantest R6245 - Dual-Channel SMU

[leighcorrigall]

Thanks for your detailed response!
Yes, I noticed the options, the datasheet even says you have to specify which one you want at the time of ordering.

Replacing the fuses shouldn't be too difficult. I noticed the transformer in one of your photos has a label "120V", which got me curious as to whether it perhaps also has a 230V or 240V tap.
Or perhaps two 120V windings that can be configured either in parallel for 120V mains or in series for 240V mains, like in e.g. the HP 3456A. The Advantest TR6142 does seem to have some similar configurability.

I'd rather not have to swap out the transformers, I'm guessing it would be quite difficult and expensive to find suitable replacements.
If you do ever take your unit apart again, could check the labelling on the transformers and whether they can be reconfigured?

They do seem like very nice instruments (which got me interested in the first place). Congrats on your score!
I might try to figure out the details of the IV clamp in the block diagram you posted.

My Advantest R8340 and R6245 both run fine even though I have a line input of 120 VAC. Odd that they never added an input voltage selector. My Advantest R6581T happens to have one.

Attached are some photos I took while I was replacing the central fan. Hopefully, they will help you figure out what you are trying to do. I do not know what I am doing with this rat's nest of wires. Good luck and let me know if they help.

The transformers can be taken out easily if you are careful and have had practice with the disassembly. They are mounted to the frame but in an awkward way. You might have to take most of the instrument apart as I have done.

The instrument is amazing and I like it so far. I will document the performance here eventually. My next task is to calibrate it. The problem is that I am waiting for calibration standard (10k precision resistor and 10 DCV source) for my Advantest R6581T, and they arrive next month.

[leighcorrigall]

Axial Fan Replacement

The original Brushless 12 DCV Sanyo Denki 109R1212H101 120x120x38 mm fan, which includes a pulse sensor, inside the R6245 is about 28 years old and still runs. However, I plan on replacing this obsolete fan with the 9WG1212E101-E from the same company having a 15 % improvement in airflow [3.34 m^3/min]. Note that the 9WG1212G101-E has slightly better performance (34 %), but I  do not know the upper current threshold of the fan circuit. To be safe, I went with similarly matched specifications and will reuse the 3-pin socket from the original.

The plan is to redirect the air intake from the rear of the unit to utilize a fan filter previously posted about. This way, I should be able to avoid cleaning the unit from contaminants. When I received this unit, it was the dirtiest thing I have ever had to clean.

The fan replacement is noticeably louder. This may offend some people, especially if you ride this down a small-town street dressed in leather. 

I suggest reversing the fan direction so that the intake is from the rear. That way a filter can be applied.

[leighcorrigall]

Attached is a CAD model of the rear bumper that was severely damaged during shipping from Korea. Keep in mind that this is one of two isomers needed to produce all four bumpers. In other words, these are mirror images of each other.

Attached is a representative picture of the printed part. There was no difficulty with attaching the part to the frame.

[chekhov]

Hi,

It's great that these transformers have all required taps to reconfigure it from parallel 100 and 120V connections into series one in order to obtain 240V. Just need to sit with paper, draw it all carefully, and understand why there is a 0 and F (internal fuse) and how they are actually connected  .

[serg-el]

Well, I'll try to help.
The scheme looks something like this.

For comparison, the scheme for reworking the transformer from  ADVANTEST TR6150.

Before

After

 

[flying_fred]

Hello everybody,

I also have this R6245 SMU for, I think, 4 years now and have already put it to good use. Many times.
It definitively is a nice instrument.

Several notes:
- It can be rewired for other main voltage (I did it for 230V voltage) and is just a matter of jumper and wires to swap on a screw terminal bloc. Unfortunately I can't remember if I took some notes about this and right now the instrument is buried under a huge pile of other instruments for a running experiment, so right now I can't take pictures and it will be some time before I can do this. Anyway it is easy enough to figure out how to rewire it.
- Please make sure to change the RAM battery on the CPU board. It is located next to the (custom IIRC) CPU and on the first SMU I bought, the battery leaks made an unrepairable mess of all this. (I was fortunate enough to then score for next to nothing the displayless version with a good CPU board)

To Leigh, I've loooong search the web, contacted Advantest, ADCMT, and some others to get some manuals, to not avail. (only the japan GPIB programming manual is available on ADCMT site)
From some of your posts I understand that you may have the (japan) user manual.
Is it a paper or pdf manual?
I would be thankful if you could post it somewhere (on ko4bb for example), that would be a great help to other owners.
And in case it's a paper manual, I'm willing to help to have it professionnaly scanned.

Thanks,
Fred

[42Khz]

Hello everybody,

I also have this R6245 SMU for, I think, 4 years now and have already put it to good use. Many times.
It definitively is a nice instrument.

Several notes:
- It can be rewired for other main voltage (I did it for 230V voltage) and is just a matter of jumper and wires to swap on a screw terminal bloc. Unfortunately I can't remember if I took some notes about this and right now the instrument is buried under a huge pile of other instruments for a running experiment, so right now I can't take pictures and it will be some time before I can do this. Anyway it is easy enough to figure out how to rewire it.
- Please make sure to change the RAM battery on the CPU board. It is located next to the (custom IIRC) CPU and on the first SMU I bought, the battery leaks made an unrepairable mess of all this. (I was fortunate enough to then score for next to nothing the displayless version with a good CPU board)

To Leigh, I've loooong search the web, contacted Advantest, ADCMT, and some others to get some manuals, to not avail. (only the japan GPIB programming manual is available on ADCMT site)
From some of your posts I understand that you may have the (japan) user manual.
Is it a paper or pdf manual?
I would be thankful if you could post it somewhere (on ko4bb for example), that would be a great help to other owners.
And in case it's a paper manual, I'm willing to help to have it professionnaly scanned.

Thanks,
Fred

A broken R6246 is on its way to me, let's hope it doesn't have the CPU board issue you described. The seller said it locks up and shows only a few @ symbols on the display.

I have an OCR'ed Japanese manual. I contacted Advantest, they told me there never was an English manual. Google translate won't eat it as it's too big, so you'd have to split it into pieces to get it through there.

[flying_fred]

Whoah, that's awesome. Thanks a million for that!
I have been looking for it for four years now... How did you get it?

Now, regarding your 6246 those symptoms were, unfortunately and IIRC, the very same that mine exhibited 
Let's get finger crossed, but be prepared to hunt for a part mule.
Again, the displayless one should be much cheaper (I believe I paid Eur90 or something for it), but lacking the "display working" information when bidding makes it some kind of a gamble. I was releived when receiving it to see the unit was perfectly OK.
It is also way more rare, at least on auction.

I expect the 6245 and 6246 share the same CPU board, apart from firmware code and the eprom is conveniently socketed. Mines were differing by some minor wire corrections but the new one flawlessly worked from start.
For my first one to be in such a bad condition it must have been left on shelf for quite some time after it failed, letting corrosion do it's nasty work. You may have better luck and your might be in much better condition.

Anyway, I still have the old board somewhere and should you need some unobtainium parts, please be my guest.

Fred.

[42Khz]

Not a problem! I got it from serg-el actually (he also posted in this thread).
I'm glad you have it now. I might see if I can pull it apart and feed it through some translator in pieces sometime.

Oh no that's not great... I had hoped it might be a display interface issue or maybe a power supply here or there.
I was considering replacing the digital guts with some modern micro as a backup plan. Although restoring the originals would be much nicer.
It was calibrated in late 2019, so let's hope the issue is very recent.

What is this displayless version you speak of? Is it an alternative model?

Thanks for your generous offer!
I'll post some picture here when my unit arrives.

[flying_fred]

When I got mine, I too was expecting for an easy fix, as it is most often the case.

Those instruments are intended for semi-fab factories (who would need that, otherwise, when most electronics design offices don't know there elbow from their arse, today? And I know quite a few...),

For lab use, as I use mine, and you probably would, the display is useful, indeed, but on a factory floor, where it will almost always be GPIB driven, the display is not needed and has been cancelled, as well all the keys. I guess more for improved reliability reasons than to cut on cost.
From what I understand they can also be used as slave units from a complete master one, to allow for complex measurement cases where lots of channels may be required.
Those displayless units are labeled R6245A (mind the final "A") and only have on front panel a GPIB address setting dip switch and a few status leds, plus the 6 triax sockets. You can find some not very meaningful mention of this in the datasheet and you can see it on page 3-9 (51) of my brand new manual  .
From what I understand there is no displayless R6246.

As for automatic translation, it might prove to be somewhat epic, as when I tried one or two illustrations titles, japanese ancestors pointed their noses out with some totally unrelated "sword" words, and so on...
Still, as I intend to GPIB program it, at least this part will need to be translated. And the 6245/46 is different enough from the R6243 I also have a pair of, and for which an english manual is available, that the 6243 manual is as unhelpful as can be.

Maybe we can setup a coordinated effort on this matter?
If some others are interested please, please tell.

Back to the instrument, I don't know how general this is a problem or not, but mine also have unbearable keyboard bounce issues, to the point that I plan to change the all switches next time I open it. (I believe those are some standard tactile ones)

[serg-el]

Back to the instrument, I don't know how general this is a problem or not, but mine also have unbearable keyboard bounce issues, to the point that I plan to change the all switches next time I open it. (I believe those are some standard tactile ones)

Surely there are the same buttons as in the R6143.
The fix is simple - the pusher is removed, the contacts are cleaned with a fiberglass stick. Aluminum sanitary tape is glued to the rubber button.

[42Khz]

When I got mine, I too was expecting for an easy fix, as it is most often the case.

Those instruments are intended for semi-fab factories (who would need that, otherwise, when most electronics design offices don't know there elbow from their arse, today? And I know quite a few...),

For lab use, as I use mine, and you probably would, the display is useful, indeed, but on a factory floor, where it will almost always be GPIB driven, the display is not needed and has been cancelled, as well all the keys. I guess more for improved reliability reasons than to cut on cost.
From what I understand they can also be used as slave units from a complete master one, to allow for complex measurement cases where lots of channels may be required.
Those displayless units are labeled R6245A (mind the final "A") and only have on front panel a GPIB address setting dip switch and a few status leds, plus the 6 triax sockets. You can find some not very meaningful mention of this in the datasheet and you can see it on page 3-9 (51) of my brand new manual  .
From what I understand there is no displayless R6246.

As for automatic translation, it might prove to be somewhat epic, as when I tried one or two illustrations titles, japanese ancestors pointed their noses out with some totally unrelated "sword" words, and so on...
Still, as I intend to GPIB program it, at least this part will need to be translated. And the 6245/46 is different enough from the R6243 I also have a pair of, and for which an english manual is available, that the 6243 manual is as unhelpful as can be.

Maybe we can setup a coordinated effort on this matter?
If some others are interested please, please tell.

Back to the instrument, I don't know how general this is a problem or not, but mine also have unbearable keyboard bounce issues, to the point that I plan to change the all switches next time I open it. (I believe those are some standard tactile ones)

That makes sense. I wasn't aware there was such a version of the instrument, thanks for the info. I've saved an ebay search in case I need it

Indeed, I gave automatic translation a quick try and what came out was fairly garbled, stuff about vacations and homes. It might turn out nearly as difficult to understand as it is in Japanese .
I guess it would be prohibitively expensive to get the manual professionally translated, even if the cost were spread over a few people.
I have a GPIB to USB adapter on the way as well, I wouldn't mind helping out in documenting out the GPIB commands. That is, if I can get my unit working.

[flying_fred]

Hi serg-el,
thanks for this, but this is not the kind of switch that we find on a 6245.

I do have a R6245, two R6243, one R6143 and one TR6142.
I never paid attention to the differences between keys (I very rarely use the 6143 and 6142), but having a check at it now I can see that both the 6142 and 6143 (same generation?) have soft rubber keys.
The 6243 and 6245, without any doubt, have the hard clicky feeling and sound of metallic dome tact switches.
What I expect is that those are not of the metallic-dome-on-PCB-pads kind that we find on some cheap mass production items, but that would be surprising.

As for the soft rubber keys, those are available on the japan yahoo auction site at a correct price (Y6000 per 30, which is about €1.50) if that helps.

[serg-el]

Okay.
Let's say they are of this type.
As in R6581.
These tact buttons should be removed. Otherwise, they don't understand well.
Flip the black rubber over. Gently wipe the round, gold-plated pad.
Clean the inside of the contacts with a fiberglass stick. 

[leighcorrigall]

How did you get it?

Sorry, I haven't looked at this thread in a while. The manual can be downloaded from https://www.adcmt.com/.

Switch to the Japanese website, not the English, and then search for R6245. You will have to fill out a page to get access to the manual. It is rather large, so I did not share it on this website.

Hopefully, this link will work: https://fo-pro.s3.ap-northeast-1.amazonaws.com/files/5d24069af875dd60ae000565/6245-series-ope-foj-8335159g00.pdf

As for an English Translation, I have already put considerable effort into doing so. Obviously, it is incomplete, but maybe we can find a platform to combine our efforts so that we can complete the task more efficiently?

English Translation:
Table of Contents
Section 2  - Before Using (partially)
Section 10 - Troubleshooting (useless information)
Section 11 - Theory of Operations
Section 12 - Calibration

[leighcorrigall]

What is this displayless version you speak of? Is it an alternative model?

There are three variations:
R6245 (channels: SMU220-2, SMU220-2)
R6245A - GPIB interface only variation of the R6245
R6246 (channels: SMU220-2, SMU62-20) high-current

[leighcorrigall]

Attached is a translation of most of Section 12 (Calibration), including figures.

Please note that I am unable to finish the "calibration of the current generation and current measurement" example code right now because my eyes hurt from looking at the screen. 

The document has been left as a DOCX (7-zipped due to website restrictions) because there might be translation errors that you can fix on your own. I tried my best to read over everything as a quality check, but I have yet to attempt a calibration.

If you find mistakes, please let me know and I will update what I have. If you want to contribute, I would be happy to share content with you here. 

どうもありがとうございました

EDIT: I removed the original upload as there is now a revision that completes the Google Translation.

[leighcorrigall]

Attached is a translation of most of Section 12 (Calibration), including figures.

Please note that I am unable to finish the "calibration of the current generation and current measurement" example code right now because my eyes hurt from looking at the screen. 

The document has been left as a DOCX (7-zipped due to website restrictions) because there might be translation errors that you can fix on your own. I tried my best to read over everything as a quality check, but I have yet to attempt a calibration.

If you find mistakes, please let me know and I will update what I have. If you want to contribute, I would be happy to share content with you here. 

どうもありがとうございました

Attached is a Section 12 update with the final code example. As I do not know the Japanese language, there will be mistakes. I have not attempted to calibrate the instrument yet and will not fully appreciate the context of the calibration procedure language.

Has anyone attempted to calibrate the Advantest R6245? If so, do you have a procedure that you can share? Thanks.

[leighcorrigall]

Well, I'll try to help.
The scheme looks something like this.

Hi, serg-el. Can you please help a colourblind noob convert this 90 - 110 VAC configured the system to a 108 - 132 VAC? I don't mind opening it up again to verify a few wires. 

From what I understand, I just need to swap some of the terminal connectors to an arrangement appropriate to the 117 VAC supply in my house.

[serg-el]

 

[leighcorrigall]

Thank you so much, serg-el! 

You are the hero we need. 

[wichers]

I implemented a pymeasure instrument driver for the Advantest/ADCMT R6245/R6246. From the 99 GPIB functions there is only a handful of commands (it's only binary/linear search & high-speed sequences) that still need to be implemented.

The (translated) documentation is quite complete already, and includes some BASIC to Python translated examples. Check it out here:
https://pymeasure.readthedocs.io/en/latest/api/instruments/advantest/advantestR624X.html

[mmathor]

Hi,

Thanks to everyone who share information about R6245/6246.
I try to translate complete Manual from Japanese language. Only some of samples in  Chapter 8 GPIB are not translated.

https://www.dropbox.com/s/fjah6dbspo23ckh/ADCMT%20R6245_R6246%20-%20Instruction%20Manual%20Rev%201.pdf?dl=0

If somebody have information about simple but powerful GPIB app or software it will be very appreciated.

Regrads

[TrexREXX]

EZGPIB
http://www.ulrich-bangert.de/html/downloads.html

[wichers]

I try to translate complete Manual from Japanese language.

Thank you for this great work!

If somebody have information about simple but powerful GPIB app or software it will be very appreciated.

Of course I am going to say pymeasure. I am by no means a python fanboy, but I really see the benefit of wrapping the GPIB commands for this instrument. The raw R6245/R6246 GPIB commands are just terrible: Booleans are all over the place (True/False are both sometimes 1 or 2), the channel parameter is not always in the same place and integers instead of enum's makes raw commands a pain.

Dual channel measurement is such a pleasure right now:

Code: [Select]

smu = AdvantestR6246("GPIB::1")
smu.reset()                                               # Set default parameters
smu.ch_A.set_sample_mode(SampleMode.PULSED_SYNC)          # Pulsed synchronized
smu.ch_A.voltage_source(source_range = VoltageRange.AUTO,
                        source_value = 20,
                        current_compliance = 0.06)
smu.ch_A.measure_current()
smu.ch_B.current_source(source_range = CurrentRange.AUTO,
                        source_value = 1E-5,              # Source current at 10 uA
                        voltage_compliance = 5)           # Voltage compliance at 5 V
smu.ch_B.measure_voltage()
smu.enable_source()                                       # Enables source A & B

for i in range(10, 60):
    k = i * 0.000001
    smu.ch_B.current_change_source = k                    # Set current from 10 uA to 60 uA

    smu.trigger()                                         # Trigger measurement
    smu.ch_A.select_for_output()
    Ic = smu.read_measurement()                           # Read channel A measurement
    smu.ch_B.select_for_output()
    Vbe = smu.read_measurement()                          # Read channel B measurement
    print(f'Ic={Ic}, Vbe={Vbe}')                          # Print measurements

smu.standby()                                             # Put channel A & B in standby