Tear it down and tell us how a precision analog to digital converter works
(especially this calibration sources with 5.5 and more digits)
*gathers up best Reagan parody voice* Mister Jones, tear down this Voltage Generator!
I made a an offer for a Advantest R6142 over a year ago on eBay, however the seller declined my offer. The unit has been on sale on eBay for over 3 years. It is sold 3 hours after Dave did a video on it. Haha, gotta love his influence.
I've got an R6871E on the way...
Just looked on eBay UK and there are a few Advantest spectum analysers for sale at 2-3K and voltage/current sources at 1.5-2K. No 300 dollar bargains! But perhaps they are rarer in this neck of the woods.
Just looked on eBay UK and there are a few Advantest spectum analysers for sale at 2-3K and voltage/current sources at 1.5-2K. No bargains! But perhaps they are rarer in this neck of the woods.
Yeah, I noticed that too. Set up a search anyway, Just In Case
Dave,. I just wanted to comment on the 2-wire vs 4-wire sensing on the R6142. It is common practice to hook the sense leads to the source leads internally with resistors. These resistors are sized so that the sensing error they add is insignificant which is why you didn't see a change in output value. But if you use 4 terminal sensing at the end of your cable, and there is a load there, you will definitely improve the delivery point accuracy by using kelvin connections. For example, let's say the sense and source leads are connected internally with 200 ohm resistors. Then try to source 100mA at 10 V through 22AWG wire. Without the sense leads, the wire resistance alone will cause a significant error. At Fluke we'd debate leaving out the 2T/4T switch as the resistors add so little error in low current measurements.
Obviously Advantest Engineers went with the resistor connected sense leads, so, for low current outputs, you would not see a difference between the 2T and 4T positions of the switch.
I'm not sure I understood the video correctly, but to me it sounded like Dave tested flipping the switch in the constant current mode -- which wouldn't even use the 4-wire connection. So nothing changing is the expected behavior in that case.
There are probably resistors as you say, so even in constant voltage mode it would probably still work, albeit possibly with a little bit of voltage error, if switched to 4-wire sensing while leaving the sense terminals unconnected.
To test the 4-wire mode you'd need a load causing significant voltage drop in the cables and also to operate it in CV mode.
Have we had a teardown of this lovely piece of test equipment yet? Have I missed it?
What's the practical use for this generator other than calibrating other test equipment?
Precision power supplies like this can be used for device characterization and design and development work. Keithley has a whole business selling precision sources like this.
Dave,. I just wanted to comment on the 2-wire vs 4-wire sensing on the R6142. It is common practice to hook the sense leads to the source leads internally with resistors. These resistors are sized so that the sensing error they add is insignificant which is why you didn't see a change in output value. But if you use 4 terminal sensing at the end of your cable, and there is a load there, you will definitely improve the delivery point accuracy by using kelvin connections. For example, let's say the sense and source leads are connected internally with 200 ohm resistors. Then try to source 100mA at 10 V through 22AWG wire. Without the sense leads, the wire resistance alone will cause a significant error. At Fluke we'd debate leaving out the 2T/4T switch as the resistors add so little error in low current measurements.
Obviously Advantest Engineers went with the resistor connected sense leads, so, for low current outputs, you would not see a difference between the 2T and 4T positions of the switch.
Adding resistors or some type of low voltage shunt is also a safety feature. If a sense lead accidentally gets disconnected, the force voltage will not rise out of bounds possibly damaging the load. Power supplies with remote sense often include this to protect their load.
I got a bargain price on a Advantest R6144 Current Voltage Generator in perfect condition
This R6144 is very similar to the one Dave showed in this video.
But it took me some time to find the operating guide / user manual but I finally got lucky.
Advantest R6144 User Manual:
ds.theatr.us/test-equipment/Advantest/6144_ope_e.pdf
If anyone wants to see some pictures from the inside of this R6144, let me know, and I will open it.
Thanks for the manual. I heard it is possible to hack R6142 into a R6144 by shorting JP1, JP2 and JP11. Maybe Dave or you can tear down your units to compare?
I also have an Advantest R6144, excellent condition, still in perfect calibration. Purchased for peanuts on eBay from a seller that had one left and was liquidating them. Excellent unit.
Thanks for the manual. I heard it is possible to hack R6142 into a R6144 by shorting JP1, JP2 and JP11. Maybe Dave or you can tear down your units to compare?
I am traveling right now, but let me take a look when I am back in the lab.
Also, I got one more manual for this R6144, may be I find the 'hack' possibility in there.
Inreresting really is how stable this unit behaves
Oh and btw, the TR6142 is the older model. Apart from the datasheet in the video saying so, at some point Advantest changed revisions in their test kit by removing the T at the start - usually the models internals were updated with this change too. I have no idea why the T was dropped off rather than change the model number!
I have found that the schematics of one revision often very similar to the newer model, an element of design reuse. So the schematics in the older TR6142 may line up in some respects.
Si.
For the general public, the "TR" models seem to have been made under brand "Takeda Riken" - and many people in Japan still refer to Advantest as Takeda.
The reason for replacing the TR with R was just the brand change.
Also, there's one more brand called "ADCMT" which seems to hold the "old" equipment, at least Adantest seems to direct queries regarding older equipment there...
Yep, it seems like any new electronic test gear is now under the ADCMT brand. There are still a few spectrum analysers on the Advantest site, but everything else is ADCMT.
I have both the TR6142 and R6142, they are certainly a little different on the inside and outside. You have to twiddle pots to adjust the TR6142, whereas the R6142 is done err... closed case style in software? (I forget the proper term =P)
Otherwise, most things from "TR" to "R" are the same.
Just a friendly warning, the Maxell backup batteries used inside most old-ish Advantest gear will need to be replaced soon, I had one explode in a nice R6145 Voltage/Current Source/Monitor, taking out quite a bit of the digital circuitry.
So don't be lazy and replace them ASAP!
Just a friendly warning, the Maxell backup batteries used inside most old-ish Advantest gear will need to be replaced soon, I had one explode in a nice R6145 Voltage/Current Source/Monitor, taking out quite a bit of the digital circuitry.
So don't be lazy and replace them ASAP!
Thanks for that info
I will take my unit apart and exchange it and at the same time take some pictures.
Thanks for that info
I will take my unit apart and exchange it and at the same time take some pictures.
I was a bit lazy because I can recalibrate the unit myself any time I wanted, so I thought it was okay if the battery dies a silent death...
Murphy got me.