Advantest R6581 8.5 digit DMM mini teardown/repair

[TheDefpom]

I have absolutely no idea how to do it yet, it is something I would have to spend a lot of time figuring out as I don't know where to sample the data, or even how the data in contstructed to convert it for display.

I was thinking I could do a video about it once I get to that point and get something working, maybe even design a PCB for it or something.

The display on mine is still working but very dim, so i'm not desperate enough to start that project yet in the hope I can one day find a replacement display instead, but I am pretty sure I will need to make something soonish.

[bsw_m]

Advantest R6581 short stability referring to JVS

[johnbaker26]

https://www.ko4bb.com/getsimple/index.php?id=download&file=Advantest/Advantest_R6581_Multimeter_Service_Manual_Schematics.rar

This link seems down. Does anyone have this they can upload?

[Ismsanmar]

https://www.ko4bb.com/getsimple/index.php?id=download&file=Advantest/Advantest_R6581_Multimeter_Service_Manual_Schematics.rar

This link seems down. Does anyone have this they can upload?

Does this one: https://www.eevblog.com/forum/metrology/advantest-r6581-service-manual/ works for you?

[johnbaker26]

Yes thank you! Though I don't see the mods that were suppose to be included in the link I posted

[TheDefpom]

Look at my videos about the unit, I did most of the mods for it, and one that isn’t mentioned to improve the input protection from high voltages

[Ismsanmar]

Look at my videos about the unit, I did most of the mods for it, and one that isn’t mentioned to improve the input protection from high voltages

That HV "mod" is mentioned on THIS thread. Please, don't start taking credit for what you haven't researched.

[TheDefpom]

Look at my videos about the unit, I did most of the mods for it, and one that isn’t mentioned to improve the input protection from high voltages

That HV "mod" is mentioned on THIS thread. Please, don't start taking credit for what you haven't researched.

What I did is NOT A WHAT HAS BEEN MENTIONED ! I did something different, did you even watch the video or did you jump straight into abusing me without checking first?

[zrq]

I think it's likely that HOSEI in the secret SCPI command is not for the name of an university but 補正, which stands for correction.

[Simmed]

https://www.ko4bb.com/getsimple/index.php?id=download&file=Advantest/Advantest_R6581_Multimeter_Service_Manual_Schematics.rar

This link seems down. Does anyone have this they can upload?

this thread ?
https://www.eevblog.com/forum/metrology/advantest-r6581-service-manual/?all

[dietert1]

The R6581T that arrived from Ismsanmar received a K2002 style mains preregulator.
Here we have about 230 Vac mains voltage and the R6581T is configured for 220 Vac. With a higher mains voltage the mains transformer and all secondary voltage regulators get pretty hot.
Images show circuit and schematic. The circuit regulates the voltage of the transfer circuit from first to second mains transformer. The rectified voltage stays at 24.5 V and remains constant to about 1E-4. Mains voltage to the first transformer is 218 Vac.
In this device i added a 2.2 KOhm load resistor to improve stability.

Regards, Dieter

[Ismsanmar]

Does that mean that the rear mains voltage selector is useless, or it is that it only configures one of the two transformers?

[Kleinstein]

Does that mean that the rear mains voltage selector is useless, or it is that it only configures one of the two transformers?

The extra mains regulator is an add-on done by Dietert. So only his unit. He likely has to set the voltage selector to 220 V. The circuit regulates the mains voltage to some 218 V (could likely be set lower), just enough to work well with the 220 V setting.
It looks like a nice way to work with stronger variations in the mains voltage, especially higher.

[dietert1]

With the 220V configuration the black and the blue wires to the first mains transformer are used and my circuit controls the blue wire. So it controls the first mains transformer, measuring its output voltage to the second transformer.
I did not look into other mains voltage configurations. Should be easy to get them working as well. If you put the point of control in between the mains input filter and the voltage configuration switch, the circuit should also work with other configurations.
Another solution could use a very similar controller only for the second mains transformer. This requires changing some part values. Maybe one would use two mosfets in order to avoid the rectifier bridge voltage drop. The circuit can have three instead of four ports.

Regards, Dieter

[Kofen]

I'll have to ask, is this really worth it? What was before and after temp? An increase from 220 to 230v equals to 5%, well below what I would consider needing any correction.

[Kleinstein]

For just the drop from 230 to 220 V it would likely not be worth the effort. The problem is more when the mains voltage fluctuates a lot, which can happen where it grid is loaded to the max. So the actual voltage may vary from 220 V to 250 V depending on the time of day.
A slight question is if the shown regulator really saves some power, or only shifts the heat from the DC voltage regulators to the added voltage stabilization. The stabilization circuit may help a little, as it can improve the power factor a little, but I would not expect a huge improvement.

[Kofen]

Yeah, you copy my exact thougths here, if the shift of power dissipation helps, I don't know, but I would expect minuscule at best. Mains are allowed to flucate at least +/- 5%, depending on region and should have been accounted for in the design. If the regs run really hot, better look at it at a lower level than trying to add a mains regulator IMO.

[dietert1]

Here i have a FLIR image that i made in October, after modding another R6581T. The transformer still is the hotspot with about 46 °C. I think it had more than 60 °C before. Don't know whether transformer losses are linear when approaching saturation.
The preregulator isn't only about avoiding heat, but about temperature and supply stability in the measurement compartment. For best precision you want to isolate the ADC from ambient as much as possible. Another example: The Prema 6048 8.5 digit meter used two voltage regulators in series, like a LM7815 followed by a LM7812. The LM7812 would then run at near constant temperature.
I'm just replacing the bad VFD by a TFT. This will also save some power.

Regards, Dieter

[Hydron]

If it's just high input voltage rather than unstable (and high) voltage then maybe consider a bucking transformer? Passive operation and very low power dissipation (just whatever the core and copper losses are from the added transformer), plus it's a heck of a lot simpler than a pre regulator.

[dietert1]

Yes, if it were only about reducing a high mains voltage to avoid heat. As i just wrote in my previous post, it isn't. Our three R6581Ts are ready with the circuit i posted above. It won't get much simpler.
Many years ago i tried this with some HP 3456A. Its LED display made supply voltages vary with displayed value and the number of lighted segments. I studied two different circuits with higher control bandwidth that would produce a clipped sine wave for the transformer, with harmonics just audible. Nine turns added into the mains transformer served as a sensor. The Keithley 2002 preregulator design i am using for the R6581T is much slower yet good enough.

Regards, Dieter

[2N3055]

Here i have a FLIR image that i made in October, after modding another R6581T. The transformer still is the hotspot with about 46 °C. I think it had more than 60 °C before. Don't know whether transformer losses are linear when approaching saturation.
The preregulator isn't only about avoiding heat, but about temperature and supply stability in the measurement compartment. For best precision you want to isolate the ADC from ambient as much as possible. Another example: The Prema 6048 8.5 digit meter used two voltage regulators in series, like a LM7815 followed by a LM7812. The LM7812 would then run at near constant temperature.
I'm just replacing the bad VFD by a TFT. This will also save some power.

Regards, Dieter

Saturation is nonlinear and so are the losses. They can be almost exponential, limited by other parameters. Also all the flux that cannot go through the core goes through anything  close by. Including causing eddy current in constructional elements (additional heating), and inducing noise into nearby circuitry.

[Hydron]

Yeah fair enough going the complicated route if the variation is enough to affect measurements. I wonder if a situation like the LED fluctuation you mentioned can't be better solved by a shunt reg though (size the dropper resistor for the all segments lit case, and the shunt reg will draw more current as they turn off, keeping the total ~constant for a given input voltage, add a series reg before the shunt to lock it down even further).

Bucking transformer is certainly what I'd use with 220V Japanese equipment - my mains supply is reasonably stable (most of the time it stays within a 5-6V range) but is pretty high at 245-250V. I should probably actually wire up a box with a 24VAC toroidal in it (got a few ex-CCTV ones) and mains connections to have handy for stuff that's fussy.

Edit: note that noise from saturation isn't just of the electrical kind - if you've got stuff that makes an annoying mains hum then try and run it at a slightly lower voltage. I'm pretty tempted to do this for my microwave actually - shame the power level is such that it's a bit tricky, plus it's almost certainly the most electrically dangerous thing you own by a huge margin.

[dietert1]

A bit OT:
These are the innards of a vintage step regulating transformer (CoNET). It was meant as a safety and noise suppression transformer for the workshop. The 300 VA transformer sits in the middle. It has two compartments separated by a grounded aluminum plate, with 4 taps of the primary that get switched automatically using triacs.

Regards, Dieter

[serg-el]

Install double conversion smartups set to 220 volts for all your sensitive equipment.

[dietert1]

We are using no-breaks (e.g. for 3D printer) and it isn't pure joy either. The switching power converters inside emit quite some electrical noise.

Regards, Dieter