Review of Microchip's new power managment IC vs other LTC one

[VEGETA]

Hi Dave, everyone...

I saw this IC on Microchip's site and it has great features that can measure true rms current/voltage and active/reactive/apparent power. The most important thing is measuring RMS voltage and current instead of doing all sorts of software sampling which I myself couldn't do it properly. It's only 2.5$ from Microchip so it's very cheap!

this is it: http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en567281

Previously, I wanted to get this one: http://www.linear.com/product/LTC1968 from Linear tech which does the RMS-to-DC conversion and thus outputs a DC voltage equal to RMS voltage of input. this one is about 4-6$ last time I checked but I could be wrong here.

Both of them works with AC voltage as explained from the datasheet, however, in MCP one it's more complex than LTC one... really didn't expect it to be like this ^_^

That is my suggestion, I think it will be a new-looking episode, right?

Looking forward to your comparison if you agreed.

Thanks to all members. 

[BravoV]

Don't speak for Dave, but gut feeling is telling me the chances he will agree is pretty slim.

I guess the audiences that have the same interest are quite low, unless he is on research and planning to do a project that needs RMS converter, also doing your request so you can save few dollars for your one off project ? 

I wish I'm wrong though.

[VEGETA]

also doing your request so you can save few dollars for your one off project ? 

I wish I'm wrong though.

you are totally wrong. Assuming that out of nowhere is just annoying... I wanted to make RMS-to-DC once and I was on short of components (not money), after that I wanted to do sampling but couldn't do it in a PIC16F877A for various reasons...

That is why I suggested this because it was a problem to me and I guess it would bother many others, I ended up calculating Vpeak of the sign wave then *sqr-root(2) to get RMS... so eventually I dodged the real problem by downgrading that design from TRUE RMS conversion to only perfect non-distorted sign wave.

I hope I cleared out that it is worth it and it is not for my on-off because that one is in the past right now and I won't be re-doing it...

[EEVblog]

Nice chip by the looks of it!
Problem with doing videos like this is, there are countless nice chips like this for all sorts of cool apps. Not enough hours in the day to looks at and play with them all :-(

[VEGETA]

I've seen these 2 chips and I think there's one from AD... don't know other chips. And, this new Microchip IC has a lot of features really, even more than LTC1968 from Linear tech which I thought it is the best.

I don't want you to look for all of them, just these 2 gives a very great idea about the operation of these kind of chips... even microchip's only will do the job, especially it is very new one! again, no need to review ALL ICs, just one is enough.

thanks anyway

[Circuitous]

I just received my demo board for it.  http://www.microchip.com/Developmenttools/ProductDetails.aspx?PartNO=ARD00455

I haven't had a chance to fire it up, but I will see if I can do that tonight.

I had a similar need, to sample voltage and current both, etc.  I was taking the isolated approach to that using the ACS756 hall effect current sensor. But I think the MCP chip will be more accurate, particularly if I use a temperature sensor as well.

I'll post back when I get some results.

[codeboy2k]

Well, it's an inexpensive part for all it can do, that's for sure; a good fit might be if you already have a design that has another micro in it already and you need this powerline monitor functionality.

But if you are doing a new design, consider the TI MSP430FExxxxx parts... The E is for Energy, this is a mixed signal part, with the full MSP430 16bit micro, an LCD controller, timers, and 3 simultaneous 16-bit ADC's.

microchip MCP39F501	(specialized, no GP microcontroller)	< $3.00
TI MSP430FE	(full micro and LCD, PWM timers )	< $8.00

The most important thing is measuring RMS voltage and current instead of doing all sorts of software sampling which I myself couldn't do it properly. It's only 2.5$ from Microchip so it's very cheap!

The datasheet shows a soft core there, likely it is a PIC with mask ROM for the calculation engine. So they are still doing the RMS calculations in software... they've just done the hard work for you here.  And in my example above, you'd have to do the hard work yourself for the TI part... I can see the advantage of someone else doing the hard work for me here

[tom66]

I have a board from a server power management system which has over 20 of these.
http://www.analog.com/en/analog-to-digital-converters/energy-measurement/ade7953/products/product.html

[Circuitous]

I fired up the eval board.  The software is okay, there's no documentation so it took a bit of fiddling to get things going.
Generally, the unit was measuring the line voltage a bit high compared to a Fluke 289 and Agilent U1272A.  The board showed 1.8 V high at 80V (on the Fluke), and 2.5V high at 120V (on the Fluke).

Overall, it would do what I need... except that I need to measure 2 loads, so two chips.

I have a board from a server power management system which has over 20 of these.
http://www.analog.com/en/analog-to-digital-converters/energy-measurement/ade7953/products/product.html

@ Tom: Those look quite interesting.  How well do they perform across the temperature range?
I see that it has 2 current inputs, can those be used to measure separate loads?  Or, just separately measure the Line and Neutral of the same load?

[EEVblog]

I don't want you to look for all of them, just these 2 gives a very great idea about the operation of these kind of chips... even microchip's only will do the job, especially it is very new one! again, no need to review ALL ICs, just one is enough.

Of course, but my point is that there are an almost infinite number of chips that could interest me, and that interest other people, and that also request that I take a look at the chip they are interested in. So I'm afraid this one isn't sufficiently interesting for me to drop everything else and do a video on it I'm afraid. I'm just so behind on videos I want to produce as it is.
So yes, it's interesting, but sorry, not interesting enough to add to my to-do list.

[electronics man]

dave has to produce videos that are of interes to as many people as posible so i dont think a chip review will be of much interest to many people