Jump to content
Thank you for the great forum, Safe Driving over the weekend. Sincerely Jason
Guest

Crit my idea ... pull it to pieces, I can take it.

Recommended Posts

Guest
8 minutes ago, plonkster said:

So you have the same problem as I do then.

The largest culprits in my house are the gate opener (0.7mA), the garage door opener (0.5mA), and the transformer of the alarm system (another 0.7mA). 

Jip, seems so.

Happened when I switched the inverter on, using the front panel. I immediately suspected that it first must "settle".
Will still test the MG when I switch Eskom power off to it.

Will remove the "porn" plugs as all that is powered on this DB board are computer related stuff and the lights ... but technically they are disconnected, as they are on a changeover switch.

Also not too fussed about it as I still run UPS'es to all the devices as I do not trust the "cleanliness" of Eskom power ...and maybe, MAYBE, because sometimes I switch things off and the wife raises her voice at me (can you BELIEVE that!) when her Pc goes off. ;-)

Share this post


Link to post
Share on other sites
Guest

A 150/35 costs about R 3 982 incl VAT.
75/15 costs R1 194 incl VAT.

So 2 x 75/15 are cheaper than 1 x 150/35 by a large margin. Keeping in mind the extra wires. But if you have spare wire ...

Now on a 24v system a 75/15 can take 440w. So 2 x 195w panels is perfect = 390w per controller.

3 x 75/15 in total = 1170w - mounting is simpler as the panels are smaller.
Ideally the new 405w panels, one per controller ... 
1 150/35 can take 3 x 355w panels - mounting is a wee bit more expensive.

Another thought: 3 x 75/15, if mounted lower than the roof, can be adjusted for winter / summer and North /East and West. 

Would you get 3 x 75/15 or 1 x 150/35?
Why?

Share this post


Link to post
Share on other sites
26 minutes ago, The Terrible Triplett said:

Would you get 3 x 75/15 or 1 x 150/35?

150/35 can be used on 48V systems, the 100/15 (or the older 75/15) is only up to 24V. So I personally went for a 150/35 so it can be as future proof as possible, but I recognise the cost aspect of it. In terms of technology, there isn't much that separates these controllers (the 150/35 is usually a rev-2 controller, but that's not too important), so for a 24V system the smaller ones might actually make sense.

SO.... Meh?

Share this post


Link to post
Share on other sites
Guest
5 minutes ago, plonkster said:

... for a 24V system the smaller ones might actually make sense.

And they can easily be sold, for 12/24v system, IF one wants to change.

Another question. Have a prepaid meter right. so I switch the feedback to grid on, want to see what is possible, for I am NOT getting the +1000w I expect.

Suddenly the panels are being USED:

image.png.e9d69cbb62830c76a57b652ec1d49790.png

 

Obviously there is no credit, BUT the power usage overall will improve.

Why not use this setting?

Share this post


Link to post
Share on other sites
5 minutes ago, The Terrible Triplett said:

Suddenly the panels are being USED:

When you turn that on, it tells the solar chargers to apply an overvoltage and it sets all solar chargers to maximum (ie the limit you set is ignored). The overvoltage is then injected into the grid.

As long as you are not billed for what you feed into the grid, and not disconnected, you could probably run like that if you wanted to.

Share this post


Link to post
Share on other sites
Guest
Just now, plonkster said:

... you could probably run like that if you wanted to.

Reminds me of a double negative? :D

I CAN run it like that as I am doing it as meter is not tripping.

Meter is a: Itron ACE9000 Relay CIU

Must just check that it is not charging me for the power. I suppose if I stand watching it feeding back, the must not be a red light flicking to show power is being used?

Share this post


Link to post
Share on other sites
Guest

Bugger, if I do send back, it does mark as usage.

So, here is the conundrum:
Do not switch on, I still pay for Eskom use.
Do switch it on, and people don't use the power, I still pay.

Which on is greater?

Here is the graph I am trying to balance, with what is:

image.png.504b716abeed88223e2c8697101a3e53.png

Share this post


Link to post
Share on other sites
Guest

Right, lets try this combination:
1) Optimize batteries - Set to 95% SOC.
2) Switch on Feed back to grid.
3) When the loads are low, use spare to charge the batts, and feed less back - play with SOC to find the balance. 

Like so: NOW the panels are working ... on the slightly cloudy day.

image.png

Share this post


Link to post
Share on other sites
Guest
9 minutes ago, plonkster said:

I don't understand the problem. 

Small array size as dictated by the <1000w average loads. Have 1330w installed - want to use that.

Looking for that balance.

So if the discharge from batts can be limited, Paul and I have logged that suggestion with Victron, then the problem of drawing 1kw+ from batts stops.

And by sharing the "research" here, shows off the power of the product, the configurability. 

Perfect world: Keep batteries charged, use the FULL power the array can give and send it to the AC loads.
I cannot seem to get that right.

Share this post


Link to post
Share on other sites
Guest

On Keep batteries charged setting, another though, if loads exceed XXXwatts, switch on feed back to grid thereby ignoring ALL limits set and "gooi klippe" from the panels.

Batts are full, so they are not bothered.

If loads drop below YYY watts, switch off Feed back to grid again, limits back in play.

That could solve it, to not feed anything back to the grid.

Share this post


Link to post
Share on other sites
Guest

Changed the feedback to grid, optimize batteries.
So batts get hammered for a few minutes are 42amps ... after having pulled 52amps from them for a few minutes. At min of 95% SOC is should not be a issue.

This is the balance I am looking for: Have not seen this from the panels ever.

image.png.9337fb2eb2825f99241bdff65a79f81e.png

And then a few minutes later ... still lekker.

image.png.75c1f9f66e15ff8290dae7f50aa18937.png

And even later .... all I need to think on, do I care if I lose units, for I'm going to pay for units, whether I use it or feed it in.

image.png.4f8d515cda591080f4a44ad85123275f.png

Few minutes later ... batts gets used ... 

image.png.1ed16122b2c140d18186970035decb51.png

Share this post


Link to post
Share on other sites
Guest

@plonkster in a nutshell, with ESS and Optimized, THIS is the part I'm trying to fine tune, where the batts are being charged, yet the load is high ... batts can wait.

Using the Keep charged, then the controllers never go to peak performance.
Switch on Feedback to grid, and then not use Optimize, you pay for the feedback ... :D

Software can deal with this.

image.png.806ae2ef3c3ee66397c101c5baeb6d01.png

Edited by Guest

Share this post


Link to post
Share on other sites

@The Terrible Triplett, fitting a type-A RCD did not fix the nuisance tripping.

So by now I have three RCDs lying around, and I also have three plug circuits on backup. And the lights can be moved off the RCD. So I can literally give each circuit its own RCD now without spending any more money. That might solve two thirds of the problem fairly easily :-)

Share this post


Link to post
Share on other sites
Guest
2 minutes ago, plonkster said:

So by now I have three RCDs lying around ...

With no technical insigth whatsoever, I had a gut-feel that it would not be easy to solve.

Refresh my memory, when does the problem actually occur your side? When you switch off the DB, and the MG takes over?

Share this post


Link to post
Share on other sites
Just now, The Terrible Triplett said:

Refresh my memory, when does the problem actually occur your side? When you switch off the DB, and the MG takes over?

When the inverter tests its relays. So basically when the power comes back after an outage, the inverter goes clickety-clickety-clack and then the RCD on the output trips.

I know it has to do with a combination of appliances and standing leakage, because it stops happening if I unplug certain appliances, but I cannot pin it down to a specific appliance. There seems to be a group of appliances, any one of them which causes the nuisance tripping, and it's either the Samsung computer monitor, the one air conditioner, the alarm system transformer, or the Telefunken television in the main bedroom, or a combination of them. I think it's a combination.

Share this post


Link to post
Share on other sites
Guest
1 minute ago, plonkster said:

When the inverter tests its relays. So basically when the power comes back after an outage, the inverter goes clickety-clickety-clack and then the RCD on the output trips

Yes, as you know, I had the exact same when I switched it on from off.

Thing is, there was a small UPS on the DB with a router and switch powered, yet the DB board tripped. If you want, it can help, I can do the same test, switch if all off, then on with nothing connected bar the DB board.

My first thought was it needs time to sync, then settle on the primary power source i.e. Eskom / Batts.

If power failures return, I'll sort it with a UPS, till you or someone, you or Victron, finds the tweak.

Share this post


Link to post
Share on other sites
2 minutes ago, The Terrible Triplett said:

If power failures return, I'll sort it with a UPS, till you or someone, you or Victron, finds the tweak.

I asked a electrical engineer in R & D about it. It seems the problem isn't that widespread...

Share this post


Link to post
Share on other sites
Guest
1 minute ago, plonkster said:

I asked a electrical engineer in R & D about it. It seems the problem isn't that widespread...

When I have a chance, I'll restart the MG with just the DB board, nothing else, and see if it trips.
If not, I'll add a device of your choice and check again? Halogen light, what would you like me to try?

Share this post


Link to post
Share on other sites
1 minute ago, The Terrible Triplett said:

what would you like me to try?

Stuff that's known to be problematic. Televisions. LCD computer screens. Some multi-plug strips. Inverter air conditioners. Apparently also some modern washing machines (those with an inverter drive). A laptop power supply with an actual earth connection.

Share this post


Link to post
Share on other sites
Guest
25 minutes ago, plonkster said:

Stuff that's known to be problematic.

Yes, but what IF the MG is switched on, just a DB connected to it, nothing else, and the DB trips?

That is what I am saying. IF that the works, the a LCD computer screen, see if it trips.

But I'm betting it will trip a empty DB.

Share this post


Link to post
Share on other sites
7 minutes ago, The Terrible Triplett said:

But I'm betting it will trip a empty DB.

Yeah, then you have a different issue. In my case, if I unplug all the downstream appliances it stops the tripping.

I need a storage scope.

Share this post


Link to post
Share on other sites
2 minutes ago, The Terrible Triplett said:

would you put 2 of these 255w panels on a 75/15 in series? 

No. At STC conditions its exactly at 75V, so at greater than STC conditions you're going to blow something. What is more, above 70V (or somewhere between 70 and 75) the MPPT will not start up. The 150/35 that I am more familiar with, for example, wants 145V or less, the 150V is an absolute maximum.

Share this post


Link to post
Share on other sites

OK, I got a little naughty again, the tool-buying kind of naughty. But let me tell the story of how this came about.

I was thinking about ways to see the leakage current, to be precise, I want to see what the waveform looks like, so I had this idea of putting a current transformer around the combined wires coming out of the RCD (so it only sees the imbalance), and then rigging that up to a storage scope.

First the idea had to be tested. I have a bunch of old CTs from a previous project, I have a spare RCD and an RCD tester, so I rigged it up as planned. Problem... 30mA is just too low to be measured by my CT (which is after all a 25A thing, it's just sized all wrong). Tried a few different burden resistors, it quickly became apparent it is not going to work. Needed a better CT.

Then I thought: A CT will only see the AC components. If we're buying stuff, then we probably want to see the DC components as well. So the next thought was a Hall-type sensor. Again, turns out those things only go down to around 100mA accuracy.

Then I thought about a shunt. Would need two, and measuring those with a scope (given the issue that the earth of the scope is... well... earthed), and that is when I thought... hang on.... surely I am not the first person to ever have this problem!?

A few minutes of googling later, and I discover (some might laugh at this!) that you get clamp-type leakage testers, complete with min/max/hold functions. And it is also just a clamp meter that you can use for other purposes. Sounds like just what I want!

(Yeah, there are cheaper ways of doing this... like splitting circuits to get the leakage down... trial and error... but I hate working in the dark, and I've always wanted a good clamp meter).

So I ordered me one of these. Expensive yes, but nothing like the Fluke. It's True RMS too (there is another cheaper model that isn't). Sadly that's only coming next week, but soon I can chase this down in scientific fashion.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...