PLEASE NOTE: If you had an account with the previous forum, it has been ported to the new Genetry website!
You will need to reset the password to access the new forum. Click Log In → Forgot Password → enter your username or forum email address → click Email Reset Link.

Notifications
Clear all

Future Feature Discussion

21 Posts
3 Users
0 Likes
450 Views
(@notmario)
Reputable Member
Joined: 3 years ago
Posts: 314
Topic starter  

Now with live data...

<a class="ipsAttachLink ipsAttachLink_image" data-fileext="png" data-fileid="1092" href="/monthly_2022_02/gauge.png.9c237c3d6d9122a7e9cb4a9e9937eb68.png" rel=""><img alt="gauge.thumb.png.d5ca803881472f010d94d95a7295c0a2.png" class="ipsImage ipsImage_thumbnailed" data-fileid="1092" data-ratio="66.80" style="height:auto;" width="1000" data-src="//content.invisioncic.com/g308908/monthly_2022_02/gauge.thumb.png.d5ca803881472f010d94d95a7295c0a2.png" src="/applications/core/interface/js/spacer.png" />

A little boring so close to idle...

Most of the rings are educated guesses. The ones on `Battery Input` use the `setup` OVP/UVP values. The more config exported in `setup` the more automatic they can be. And safety alarm/shutdown values can be used as well.
Is outPF functional? I presume it meant power factor.


   
ReplyQuote
(@sid-genetry-solar)
Member Admin
Joined: 4 years ago
Posts: 2890
 

Had a response written up this morning...then the WiFi at Sean's went down right before I clicked the "Submit" button--and my reply went into the digital circular file...

 

Posted by: @notmario
So, compressed i actually meant gzip (non minimized, additional minimization would probably get you sub 1kB). You might be aware, pretty much all standard web clients support gzip. What you would do is store the JS in gzip and you could either: have the web server set content-encoding:gzip, and just send it as-stored (the browser decompresses for you). Or you can implement libz and decompress when serving the file. The latter option can of course use any decompression technique you want - but of course the decompression functionality, itself, uses space, though a lot of embedded environments will already have it built in. Incidentally, you can do this with anything served by the web server - so all assets can take advantage of it - big savings to be had.

Leave it up to me to not be aware of such a very simple thing--this could come in extremely handy!

I did test it on the Logo.png--only to find that GZipping it made no difference whatsoever to the size.  (Regular Zipping actually makes it larger.)  However, a GZip on the main HTML file drops it from ~11kb to 3.4kb--now that's an improvement!

Thanks for the suggestion.  Definitely very valuable...

 

Posted by: @notmario
Most of the rings are educated guesses. The ones on `Battery Input` use the `setup` OVP/UVP values. The more config exported in `setup` the more automatic they can be. And safety alarm/shutdown values can be used as well.

Absolutely correct that I should export more of the settings to "setup" for gauges/info 😉.  Goal of the gauges on the inverter local server Status page is simply to "beautify" the raw data provided by the inverter--data that could be inputted into an Arduino project or otherwise. 

 

Posted by: @notmario
Is outPF functional? I presume it meant power factor.

So....your guess is correct.  Answer to the question is more of a "yes and no"....

In the A.1 / B boards, total output power was monitored and calculated by a specialized power monitor IC.  (Which conveniently is basically completely out of stock in the U.S. at the current time).  These board setups did have Power Factor--it was simply another of the 2 dozen odd calculation outputs that the chip provided.  Trivial to put it on the screen / output data.

On the Rev. C boards, total output (and now input) power is measured by the main CPU--which provides a MUCH higher throughput and reaction time (as the power monitor IC ran 3x/sec).  I don't think the CPU has enough processing power to calculate power factor--for what little use it is these days anyway.  Most of us don't care too much about how the appliance is gobbling power--we just want it to run.  So...for a Rev. C setup, the outPF is currently unimplemented.

 

 

It appears to me that the tick marks are manually placed, not calculated by position?

 

Only other comment (besides a tad more styling just for my picky eyes)...is it easily feasible to use an SVG "pointer" image or something for the pointer?  Something a tad "shapier", yet without having to calculate complicated polygon points...I don't want to do THAT math!


   
ReplyQuote
(@notmario)
Reputable Member
Joined: 3 years ago
Posts: 314
Topic starter  

No surprise that GZIP couldn't compress the PNG. PNG uses DEFLATE compression internally. GZIP is an "evolution" of the same DEFLATE algorithm. Nothing more for GZIP to do.
Yes, GZIP works very well on any kind of code-ish content.

The tick marks are calculated the same way as the needle. As are the numbers. (The ticks and numbers are technically independent, though humans prefer to line them up with each other) They are however manually defined, allowing you to put them at human-sensible locations rather than mathematically assumed ones. I don't generally think of that as an issue since min-max ranges tend to stay steady. A wrapper function can auto-fill based on some math, of course.

This is how a gauge is currently configured:

<div class="dc volts gauge inputs BattV" title="Battery Input
(Vdc)">
  {
    "range": {"min":0,"max":60},
    "rotation": {"min":-45,"max":225},
    "ticks": [0,12,24,36,48,60],
    "numbers": [0,12,24,36,48,60],
    "ring": {"radius": 88, "segments":{}},
    "json_params": {
    	"ring": [[0,"red"],[["setup", "UVPe"],"yellow"],[["setup", "UVPa"],"green"],[["setup", "OVPa"],"yellow"],[["setup", "OVPe"],"red"],[60,""]]
    }
  }
</div>

It's a little hefty, but it works... The "yellow" etc are user-defined CSS classes that can be styled however - gradients, etc. The rings are actually calculated SVG circles.

As for the pointer. Yes, SVGs can be used. Simply modify the appropriate definitions in the stylesheet. "Drawn" things are outside my wheelhouse, though.


   
ReplyQuote
(@notmario)
Reputable Member
Joined: 3 years ago
Posts: 314
Topic starter  
On 2/8/2022 at 8:28 PM, Sid Genetry Solar said:

I'm hoping for mid-80s out of the current iteration (i.e. typical run of the mill).  Don't have any actual measurements, though...

A 10% difference in efficiency is hugely significant in the heat generated in the realms these inverters run (80% efficient will generate DOUBLE the heat of a 90% efficient unit).

On the GS...

I'm calculating ~87.5% at ~5000W load.
~77% at ~300W.

I would like to get you overall average efficiency over many days... but i've forgotten when i reset the KWH meter. 😕
So... I'll reset it now... mark the time. 🙂


   
ReplyQuote
(@sid-genetry-solar)
Member Admin
Joined: 4 years ago
Posts: 2890
 
9 hours ago, NotMario said:

On the GS...

I'm calculating ~87.5% at ~5000W load.
~77% at ~300W.

I would like to get you overall average efficiency over many days... but i've forgotten when i reset the KWH meter. 😕
So... I'll reset it now... mark the time. 🙂

300W output @ 77% = 390W in = 90W loss.  Is the inverter running either of the cooling fans at this load level?  (87.5% would be 43W loss--so I guess there's only room for ~35W of fans, haha!)

87.5% @ 5kw is I guess not too surprising 😉.  Kinda wish it was higher--but the only way to increase efficiency is to increase unit cost and weight (materials + shipping) by beefing up the transformer windings a bit.

I'm pretty sure I can write code for the inverter to determine realtime transformer efficiency.  Might be an interesting metric to see...and if theories hold up straight, it should be pretty constant over the load range.  (Lower loads obviously will run lower total system efficiency, as the inverter base power requirements + cooling fans are considerably more significant in lower loads.)


   
ReplyQuote
(@notmario)
Reputable Member
Joined: 3 years ago
Posts: 314
Topic starter  
Posted by: @sid-genetry-solar
300W output @ 77% = 390W in = 90W loss. Is the inverter running either of the cooling fans at this load level? (87.5% would be 43W loss--so I guess there's only room for ~35W of fans, haha!)

Inverter is in 30f ambient. The fans don't run [at all] until the load is pretty high. (Like 2KW...) Even at 5KW, it was only at 45% fan.
At such a low load i don't really find it surprising it would be in the 70's. Most other inverters are far worse because of the base idle draw. (Which doesn't go away with load)

88% peak is, IIRC, what PJ advertises. I've seen a few inverters claim 92%. AIMS advertises 88%.
I'd say you're competitive at minimum.

Worth noting that this is a 24v system, not 48v. There's probably 1-2% efficiency reduction right there.

From what i was seeing last night, the efficiency was around 86-88% from 2KW through 5KW. From 300-1500W-ish, the efficiency increases rapidly from 76%ish to 88%ish. Like i said, the fans don't turn on until around 2KW or more, so that should not be a factor at those low numbers.


   
ReplyQuote
Page 2 / 2