The transformer is mounted on the side and on an in-dented plate to give it the most robust mounting arrangement. Flat plates bend, throw in some angles and the strength increases.
As you said, for DMMs like this higher sample rate trades off for less accuracy. But it can be useful for high speed measurements sometimes, more accurate than a scope. Guessing they limited the sample rate based on processing or something else. Good teardown.
It is. Precision custom metal film networks, TCR down to 1ppm, stability class 0.025%. The Max6225 datasheet contradicts itself - the 1.5ppm typ, 3ppm max applies to the DIP package whereas the AESA+ (SO package) is shown as max TC of 5ppm in the table on the first page. Who knows what the typical value is supposed to be? (Not that 'typical' means very much).
The use of Fujicon capacitors, Analog Devices op-amps, and the Max 6225 voltage reference really speaks to the commitment to quality. Especially impressed with the custom input connector design and the robust input protection circuitry. Unfortunately here in Brazil, such an equipment would cost a kidney. 🥲
"Entering the high-range category"..... I do agree with you, seeing the quality of the build, but it becomes also barely affordable for a hobbyist.....
Bench top multimeters are generally not the right choice for hobbyists who only use them occasionally and they don't need the precision or the accuracy of such an instrument. There are hand held meters that are more affordable for that type of usage.
to be honest Unity was and is always cheap and crap ..for expensive models. they are not crap but they are not special in any way so from all those close ups..it looks pretty bad
@@sebastian_harnisch Hello. Yes i can share few details. 1. Cooling of those heatsink is not needed. Instead of small heatsink they could put bigger one or use chassis as heatsink as it is done with other models or they could use specific low dropout regulator 2. Those crimps/input terminals looks skimpy at best plastic doesn't look like any good choice..yes it is cheap but that is about it.. ..i didn't see any good precision resistor there..yes there were few but for good equipment you need good low tempco resistor with good voltage source. 3. Voltage source is so so ..nothing special i would rather use heated voltage standard instead of just solution on chip but this is just my discussion. 4. RF shielding isn't good as i see it.. i didn't notice imput dedicated chassis filter(but i might be wrong on this one). ..and much more.
@@SylwerDragon I find it a bit difficult respond. I guess it depends on what you expect. You’re neither paying for Keysight nor are you getting Keysight. It’s not a reasonable approach to just say „you need this or that number of precision resistors. You have to characterize the unit and see whether it meets its specs. If it does this argument is pretty pointless. If it doesn’t then *that* is a problem. I agree that it would be nicer to have a voltage reference like the LM399, but then again, for an affordable 5 1/2 digits DMM the choice seems somewhat reasonable to me. I’m not 100% sure, but I think that most direct competitors (Siglent, Hantek etc.) have a very similar approach (MAX6325 or similar). If you know of an equally affordable new DMM with a better voltage reference, please let me know. I‘ll leave it at that. This is not to say that I‘m 100% impressed and would recommend to blindly buy cheap test equipment - and personally chose to go with two used 34401A. But I think it’s important to state facts and give explanations so that one can judge whether an argument is worth considering.
@@sebastian_harnisch I understand.. to your findings lets say. 34401A is what i bought twice myself including some improvements i made on it and such..but that is basic of quality/design i would start with 6.5Digit. Ofc i understand cheaper you go less options you will get..but again ...As company you don't care about cheap equipment you pay a bit more for good one ..but that will last for years to come.. as individual used equipment is way to go..but there are brands you don't want to go with and UNITy is one of them but ofc choice is yours as they say :) not because they make bad equipment but because they usually make cheaper equipment with less quality/design on mind and more to price.
The transformer is mounted on the side and on an in-dented plate to give it the most robust mounting arrangement. Flat plates bend, throw in some angles and the strength increases.
Makes sense maybe from some weird manufacturing limitation because why wouldn't they be able to the same indent to the bottom metal plate?
Feel free to send me that old glitchy unit when you get the new one. I don't mind putting up with it 😊
Matt.
That's a very valid point for those that are going to comment and trash the model or mfr because of something like that :-)
As you said, for DMMs like this higher sample rate trades off for less accuracy. But it can be useful for high speed measurements sometimes, more accurate than a scope. Guessing they limited the sample rate based on processing or something else.
Good teardown.
Thank you!
do you sell your multimeter?
7:04 The logo on that UPRNS resistor network looks like it might be the logo of Thunder Precision Resistor.
It is. Precision custom metal film networks, TCR down to 1ppm, stability class 0.025%.
The Max6225 datasheet contradicts itself - the 1.5ppm typ, 3ppm max applies to the DIP package whereas the AESA+ (SO package) is shown as max TC of 5ppm in the table on the first page. Who knows what the typical value is supposed to be? (Not that 'typical' means very much).
Thank you for sharing that. Again, another valid point for me saying a lot of high quality parts seen in this assembly.
The manufacturer of these voltage dividers is a company TOKEN [Token Electronics Industry Co., Ltd.]
Nice! Thank you for sharing.
The use of Fujicon capacitors, Analog Devices op-amps, and the Max 6225 voltage reference really speaks to the commitment to quality. Especially impressed with the custom input connector design and the robust input protection circuitry. Unfortunately here in Brazil, such an equipment would cost a kidney. 🥲
"Entering the high-range category"..... I do agree with you, seeing the quality of the build, but it becomes also barely affordable for a hobbyist.....
Bench top multimeters are generally not the right choice for hobbyists who only use them occasionally and they don't need the precision or the accuracy of such an instrument. There are hand held meters that are more affordable for that type of usage.
About the same as rigol and siglent in this segment. Something is better, something is worse. But the general solutions and components are the same.
I generally agree with that view and important to add is that UNI-T price their models lower than the competition.
to be honest Unity was and is always cheap and crap ..for expensive models. they are not crap but they are not special in any way so from all those close ups..it looks pretty bad
What looks pretty bad exactly? Can you be more specific?
Please share more details, it's important to understand what looks pretty bad to you.
@@sebastian_harnisch Hello. Yes i can share few details. 1. Cooling of those heatsink is not needed. Instead of small heatsink they could put bigger one or use chassis as heatsink as it is done with other models or they could use specific low dropout regulator 2. Those crimps/input terminals looks skimpy at best plastic doesn't look like any good choice..yes it is cheap but that is about it.. ..i didn't see any good precision resistor there..yes there were few but for good equipment you need good low tempco resistor with good voltage source. 3. Voltage source is so so ..nothing special i would rather use heated voltage standard instead of just solution on chip but this is just my discussion. 4. RF shielding isn't good as i see it.. i didn't notice imput dedicated chassis filter(but i might be wrong on this one). ..and much more.
@@SylwerDragon I find it a bit difficult respond. I guess it depends on what you expect. You’re neither paying for Keysight nor are you getting Keysight.
It’s not a reasonable approach to just say „you need this or that number of precision resistors. You have to characterize the unit and see whether it meets its specs. If it does this argument is pretty pointless. If it doesn’t then *that* is a problem.
I agree that it would be nicer to have a voltage reference like the LM399, but then again, for an affordable 5 1/2 digits DMM the choice seems somewhat reasonable to me. I’m not 100% sure, but I think that most direct competitors (Siglent, Hantek etc.) have a very similar approach (MAX6325 or similar). If you know of an equally affordable new DMM with a better voltage reference, please let me know.
I‘ll leave it at that.
This is not to say that I‘m 100% impressed and would recommend to blindly buy cheap test equipment - and personally chose to go with two used 34401A. But I think it’s important to state facts and give explanations so that one can judge whether an argument is worth considering.
@@sebastian_harnisch I understand.. to your findings lets say. 34401A is what i bought twice myself including some improvements i made on it and such..but that is basic of quality/design i would start with 6.5Digit. Ofc i understand cheaper you go less options you will get..but again ...As company you don't care about cheap equipment you pay a bit more for good one ..but that will last for years to come.. as individual used equipment is way to go..but there are brands you don't want to go with and UNITy is one of them but ofc choice is yours as they say :) not because they make bad equipment but because they usually make cheaper equipment with less quality/design on mind and more to price.