ADE7756 vs CS5463

Go To Last Post
7 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi, gentlemen:

I'm evaluating the possibility to add one of the two IC's mentioned to one of my projects, to analize 'power quality'.

As far as I digged into the datasheets, it seems to me that CS5463 gives more information and prestations than ADE7756, while the latter has some outputs that allow an 'stand alone' working options.

It also seems that ADE is more like an 'energy counter' while CS is a 'power quality measurement' sytem.

Has someone here experience with them? Comments? Suggestions?

Thank you in advance,

Guillem.
"Common sense is the least common of the senses" Anonymous.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I used the ADE7754 in a 3-phase power monitoring app fully "loaded" with CTs and voltage dropping. It was not an "energy metering" app at all, but rather "motor protection".

I found that I didn't even have to touch much of the calibration and fine tweaking that may be done with a billing application. In most cases the 24 bits was overkill and I dropped the low 8--16 bits of voltage level or current was more than enough for the industrial app.
http://www.anderson-bolds.com/Me...

We have also applied '7753 and '7763 in some test stands and specialized monitoring apps. Given the familiarity with the '7754 and faced with an app (for example) to measure/display/act on the phase difference between the AC current and voltage waveforms we applied the ADE chip though it was way overkill. Given the target market is energy meters that are produced in quantity, we have the advantage of being able to get a few at a very attractive price.

Hmmmm--how come the '7756 does not show up on the table at http://www.analog.com/en/subCat/... ? Aaahhh--it is becoming obsolete; the product page indicates "Last Time Buy".

I'll dig into the CS5463 to see what you mean, but the ADE chips with the SPI interface had MORE information than I'd ever use.

A quick pass through the datasheet at the Cirrus site seems to indicate that they are essentially quite similar to me. I'd say they were "direct competitors". What extra pieces of information does the Cirrus have?

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thank you very much, Lee. This a really useful post, with lots of information.

In brief, we also don't want to use it to bill any energy, but the main use will be double: motor protection in single phase (similar to your posted project, very nice, BTW), and also power consumption/energy metering per hour.

It is intended as a 'gadget' to add functionality to a controller that takes care of a certain machine, that have few different electric motors and different setpoint/adjusts that can affect overall energy saving. Thus not only protect, but also add energy saving functionality.

For a bigger application, where 3-phase motors, we are analyzing the possibility to do all this with a 'number cruncher' that also measure harmonics and waveform distortions, but the idea to use this IC also for a 3-phase is appealing. Can you tell me how do yo do that in 3-phase? Aaron configuration or three different IC's?

The main difference, from what I could see right now between both, is that while ADE has an internal register that counts energy, the Cirrus IC don't have this register, thus one needs an external counter. On the other hand, the CS5463 measures waveform distortion (in power, not in V neither in I), thus no need to do any FFT.

Of course, for harmonics calculation and phase imbalance-assimetry-out of phase in 3-phase systems, there is no way to get rid of FFT.

Anyway, we are still in the study phase, so still not shure about wich exact features we should give to this products. So this is the main reason why I can't tell more about the intended application.

I will also take a look at '7753, or even better, to take a deeper look at AD's web page. Since I'm using some AD's ADC's, I'm quite confident in their quality. That is why my first chance would be their IC, but this depends also on the final requirements.

Thanks for you help.

Guillem.
"Common sense is the least common of the senses" Anonymous.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Can you tell me how do yo do that in 3-phase? Aaron configuration or three different IC's?

neither. Use the ADE7754 or one of the other 3-phase products. That is what is in the device I linked to.

Quote:

waveform distortions,

Note that there is already some of that--at least the waveform capture--in the ADE.

Quote:
measures waveform distortion
If that ends up to be important then that may sway your decision. ;)

The ADE chips are readily available--most in-stock at DigiKey--and not expensive. It does take a bit of time to set them up because there are so many configuration registers. When I did my second app and had the more-complicated 3-phase '7754 code as an example it took a day or two to get it going. The first, 3-phase app, took maybe a week to make a "driver".

I'd suspect that the Cirrus would be about the same. Try them both.

Analog Devices was very helpful when we did the '7754 design. The chip was quite new at the time so they were monitoring users/design wins quite closely.

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the information, Lee. I still have to take a look at this IC's. Right now, I have a bunch of meetings, 'brainstormings', few documents to finish off, and few other tasks to finsh from other projects that are about to be released (or similar :S). Probably next week I cold take a deeper look at them. I don't know when will be ready to test this chips (and many more), but I expect about Q4, this year.

Right now we are only 'throwing ideas', with few hints about what is in the IC market, and past summer we will probably go further investigating all possible lines. But the fact that AD have some 3-phase products is a major point to us. Probably Cirrus also have some.

The main problem with waveform distortion is that certain motors run warmer, and in our case, this could be a problem, while for resistors and fans, this is not an issue.

Anyway, we still should investigate what kind of electric disturbances are the most problematic, and what protections should we implement. We would worry later about how we implement them. Overheating motors can be protected with a simple temperature probe, or even with a thermostat.

I like AD products. I worked with AD7708 16 bit ADC, also SPI, and it tooks me few days to interface it and make a "driver". It works like a charm, although it has a lot of registers inside. The waveform capture in the ADE would be also a good point, since we can use it as a simple external ADC and perform any waveform analisis through that. The fact that we need a wide dynamic range for the current (from few mA to, perhaps, 20A) to work with the different loads that we would handle is also a good point, since 12 bit would not be enough (and AD7708 would be too slow).

Respect to the price, it seems that both are 'cheap', in the sense that the main cost would be isolate them and other electrical issues. And if we take into account the fact that to do a similar job one needs an external ADC with 16 bits or more (due our dinamic range in current, probably in voltage also), then the price difference becames even smaller.

A pity that this project probably would not be AVR powered (neither Xmega :(). Perhaps certain MIPS based uC? This decission is not in my hands.

Guillem.
"Common sense is the least common of the senses" Anonymous.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

The main problem with waveform distortion is that certain motors run warmer, and in our case, this could be a problem, while for resistors and fans, this is not an issue.

Anyway, we still should investigate what kind of electric disturbances are the most problematic, and what protections should we implement. We would worry later about how we implement them. Overheating motors can be protected with a simple temperature probe, or even with a thermostat.


So you are trying to elbow into the market for the product that I linked to above?!? There goes my royalty checks. :(

Note that the MPA already has provision for temperature monitoring and tripping. ;)

Wait till you try to implement the IEC overcurrent profiles, trying to act like a melting fuse with the high-precision digital samples.

Will you handle 10kV ("medium voltage") like the MPA?

In my limited experience in working with test sites for that device, no-one ever even asked about distorted waveforms. But they >>were<< amazed when they learned how imbalanced from phase-to-phase their pet motors really were. (the motors, not the supply)

Quote:

And if we take into account the fact that to do a similar job one needs an external ADC with 16 bits or more (due our dinamic range in current, probably in voltage also), then the price difference becames even smaller.

Yes, I do not think that you can match the cost of building an equivalnet. And to get equivalent capabilities would take much engineering. Heck, they are only about $2 in qty. 100 from DigiKey, and in-stock to boot.

Quote:
Perhaps certain MIPS based uC?

I can see that if you really, really want to do waveform distortion in real time then you might want processing power. But there is soooo much stuff already inside the ADE7754 that we didn't even touch for the app above. Mega32 was used, and speed is no problem. The information is gathered every zero-crossing of one phase and then a number of calculations are done for trip points and other alarms. That takes maybe 10% to 25% of the AVR running at 7.3728MHz. Add some display tasks and remote commo and logging and it is still lightly loaded.

Heck, you've already got the waveform capture. What difference would it make if an AVR chunked away and gave an answer every few seconds instead of every few milliseconds? If the waveform >>is<< distorted what can you do about it? [I suppose you could be working on an inverter system.]

Lee

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Je, je. Don't worry Lee. We wouldn't be competitors. I can't post publicly what we are working on, but definitively is not a similar system. And neither an inverter.

We do controls that would control a whole 'big' machine, that includes few lights, fans, resistors, motors. And since we control all this, we would offer the possibility to gather information and supervise / protect all this stuff. This would be a 'ready made' electric panel where the final customer only has to plug the power cords of the different parts, and done. And with certain control routines that reduce the electrical bill ('energy saving' in this days that fuel is not cheap is a good point).

And since we hardly sell to other european countries, I would be difficult that we sell anything in the US ;).

My first guess for the uC was to use the little PSoC, that would be enough if 'waveform distortion' were not needed. but the response was that '32 bit is a good marketing reason' to use bigger uC. Even if PSoC fits the bill.

Well, better if I don't explain all the marketing stupid things and thinks that I heard, otherwise I will get depressed :(

Guillem.
"Common sense is the least common of the senses" Anonymous.