navitron
 
Renewable Energy and Sustainability Forum
Welcome, Guest. Please login or register.

Login with username, password and session length
News: Anyone wishing to register as a new member on the forum is strongly recommended to use a "proper" email address - following recent spam/hack attempts on the forum, all security is set to "high", and "disposable" email addresses like Gmail, Yahoo and Hotmail tend to be viewed with suspicion, and the application rejected if there is any doubt whatsoever
 
Recent Articles: Navitron Partners With Solax to Help Create A More Sustainable Future | Navitron Calls for Increased Carbon Footprint Reduction In Light of Earth Overshoot Day | A plea from The David School - Issue 18
   Home   Help Search Login Register  
Pages: 1 ... 4 5 6 7 [8] 9 10 11 12 ... 20   Go Down
  Print  
Author Topic: Grid Hopping  (Read 61056 times)
Scruff
Hero Member
*****
Offline Offline

Posts: 1230



« Reply #105 on: February 25, 2017, 01:18:36 PM »

The trick is to be very expensive those rare days you do actually go to work Roger.  Month on, month off. Wink

UV leds didn't work. I'll try the sun whenever that comes back.  fingers crossed!
Logged
Scruff
Hero Member
*****
Offline Offline

Posts: 1230



« Reply #106 on: February 26, 2017, 05:51:35 AM »

I disassembled the PV tanning bed in the man cave and sher seeing as there was a PSU up there now I installed some under-counter lighting.



Later I had a look at this PWM -> GTI changeover circuit.
I got this relay for the task and drew the pinout config.



It's a double pole double throw relay rated 10A @ 30VDC and 5A @125VDC with a 12v coil.



I decided to series the switches to reduce the risk of arcing, switch wear and reduce the overall current per contact.



{don't mind what's under the paper scrap...I was having notions of switching the negative..but decided against it because I knew it would bite me in the bottom eventually}

Note on mains (derived) / battery changeover control circuits; Always power the relay coil from the mains (abundant) side.


Here's the ingredients I had to hand.



The relay holder has soddin' captive screws.. fume which meant I'd have to use fork terminals, which I hate cos they're garbage and I don't have any for those two reasons.
They're a liability they rattle lose fall out and make smoke. Ok so houses don't vibrate anywhere near as much as mobile installations...in that aspect houses are way easier to wire. All the same if you have a lose enough termination the PWM wave can untap it.
That left me no better option than butchering some ring terminals that would've done a great job if ease of install wasn't a greater concern to the relay designer than safety in the field.  Angry



Not having an MC4 crimper I went about soldering the tails for the PV throughputs from the relay DIN rail holder.





They're pretty forgiving terminations..you don't have to remember to put the compression gland on first.

On the other end of things I put a photo sensing relay in an IP55 enclosure on M3 x 10mm PCB standoffs with the LED protruding.



I reckon it's sealed as much as a 1.50 component demands (the enclosure probably cost more than the device  Roll Eyes).



Meanwhile back at the ranch..as you can see I changed the worklight to the magenta-y - white.. Tongue



I mounted the relay and connected it up. I needed the three way MC4 to give me clearance out the side of the Soladin for a ground for the TriStar...there's only 2 ways connected though.

cr@p meter still isn't calibrated to zero vots (world of pain getting that contraption to register...best leave it alone...if I eff it up I'll be days wishing I was only 1 volt off... banghead)
I will of course put that relay in an enclosure after prototyping with panel mount MC4s...for now I'm trying to get a sense of the mechanical layout of the finished power plant...as much as I can I want to keep mixed voltages separate but at the same time not everything needs it's own dedicated enclosure.

These things even have reverse polarity protection.  whistlie




Changing of the guards...yurp no capacitor bus inrush...much better for all involved. So PWMing a 100V array onna 48V battery is kinda making 40% of it redundant...meh...float charger and better than importing fossil lecky for my GTI to think about doing some graft. Anyways it'll be a 60V array when I'm done I'm just catering for the Soladin atm.




Ready for external mounting of the sensor tomorrow, calibration and trials.  Smiley
  



« Last Edit: February 26, 2017, 12:06:41 PM by Scruff » Logged
Iain
Hero Member
*****
Online Online

Posts: 1721


« Reply #107 on: February 26, 2017, 10:04:10 AM »

Hi

Quote
.I decided to series the switches to reduce the risk of arcing, switch wear and reduce the overall current per contact.

Don't understand that. Surely as the contacts are in series, the current will be the same through each contact.

If you have them in parallel the current will be reduced.

If the PV is on 9&12
GTI on 1&4
PWM on 8&5
Iain
« Last Edit: February 26, 2017, 10:07:41 AM by Iain » Logged

1.98kWp PV (11 x Sharp 180 and SB 2.5)
20 x 65mm Thermal and 180ltr unvented
Powervault 4Kw - G200 Lithium-Ion (LiFePO4)
9000ltr rainwater storage   Plymouth
Tinbum
Hero Member
*****
Online Online

Posts: 1177


« Reply #108 on: February 26, 2017, 10:59:38 AM »

Hi

Quote
.I decided to series the switches to reduce the risk of arcing, switch wear and reduce the overall current per contact.

Don't understand that. Surely as the contacts are in series, the current will be the same through each contact.

If you have them in parallel the current will be reduced.

If the PV is on 9&12
GTI on 1&4
PWM on 8&5
Iain

Its done in all of my PV DC breakers by the manufactures. The more in series the higher the voltage rating.
Logged

85no 58mm solar thermal tubes, 28.5Kw PV, 3 x Sunny Backup 5048, 3x Sunny Island 5048, 2795 Ah (135kWh) (c20) Rolls batteries 48v, Atmos wood gasification boiler, Brosley wood burner, 2000lt buffer tank and 250lt DHW
Scruff
Hero Member
*****
Offline Offline

Posts: 1230



« Reply #109 on: February 26, 2017, 11:48:54 AM »

Hi Iain,

It's the disconnect that concerns us the most with switching DC. Putting it in series makes a cascade albeit a near instantaneous one. What happens is the arc upstream is extinguished by the open circuit downstream or vice versa. If they were parallel the load carrying capacity would be halved this is true but the breaking current would be subject to the slowest contact.



I did describe that badly, what I meant by the overall contact current being reduced I meant in a breaking scenario. They both handle full load when made.

I've mentioned elsewhere previously how I avoid manufacturer metering where-ever possible...because they omit, fudge, forget, deceive, offer peace of mind, are liberal with the truth etc...
Here's a shining example of that.



Meter on the LHS cost me 15 and tells the truth.
Meter on the RHS costs 90 and lies.  police

Do people not want to know it takes days to reach true zero current?
« Last Edit: February 26, 2017, 12:09:27 PM by Scruff » Logged
jonesy
Hero Member
*****
Offline Offline

Posts: 839



« Reply #110 on: February 26, 2017, 02:46:57 PM »

Just have a squint again at the datasheet for that mh2.  Omron isn't alone in putting the approved power ratings towards the back, with great headline figures at the front. So whilst they state 110v/5A, they don't mean at the same time.  It's true the relay will take 110vdc, but on the endurance curves they don't go above 24v. The tuv rating does not permit dc at all, American 30v. You're in series, but you're well beyond 30v per contact.
Suck it and see, but don't put too much money on it.

You don't want my French wine. The diesel is cheaper and tastes better. Currently paying 1.30/l for French red, about 1.10 for Spanish wine mountain.  Hangover is much worse on spanish.  White is 80c a bottle and the glass is worth more. Hangover starts with the cork pull.
Logged

1.1kWp PV & SB1700. 7kW log burner.
Scruff
Hero Member
*****
Offline Offline

Posts: 1230



« Reply #111 on: February 26, 2017, 03:00:13 PM »

Yurp I flagged that alright, when I saw 30VDC on the relay with the Chinese Export brand. Had my suspicions without pursuing it as far as you did...both RS and Farnell are listing it as 125VDC.  fume
(Holder was 5 cheaper in RS)
It might be a sacrificial component the operational design is to switch at 30W or 600mA so I'll probably be ok.



Of course it's observations like that I'd have you over for a chinwag any day Jonesy. ballspin
Logged
Iain
Hero Member
*****
Online Online

Posts: 1721


« Reply #112 on: February 26, 2017, 03:32:31 PM »

Hi Scruff

All clear now, it was just the way it was worded, Thanks


Hi Tinbum
Quote
Hi

Quote
.I decided to series the switches to reduce the risk of arcing, switch wear and reduce the overall current per contact.

Don't understand that. Surely as the contacts are in series, the current will be the same through each contact.

If you have them in parallel the current will be reduced.

If the PV is on 9&12
GTI on 1&4
PWM on 8&5
Iain

Its done in all of my PV DC breakers by the manufactures. The more in series the higher the voltage rating.


Yes the more in series the higher the voltage rating, but he said, series to:-
Quote
reduce the overall current per contact.

depends what you are trying to reduce, voltage or current
Parallel reduces current/contact
Series reduces voltage / contact

Iain
« Last Edit: February 26, 2017, 03:36:19 PM by Iain » Logged

1.98kWp PV (11 x Sharp 180 and SB 2.5)
20 x 65mm Thermal and 180ltr unvented
Powervault 4Kw - G200 Lithium-Ion (LiFePO4)
9000ltr rainwater storage   Plymouth
Scruff
Hero Member
*****
Offline Offline

Posts: 1230



« Reply #113 on: February 26, 2017, 05:21:38 PM »


Yes the more in series the higher the voltage rating, but he said, series to:-
Quote
reduce the overall current per contact.

depends what you are trying to reduce, voltage or current
Parallel reduces current/contact
Series reduces voltage / contact


Indeed...best not assume I always know what I'm talking aboot.  Grin When you break it though at least to my mind they are both. Two switches are disconnecting in parallel albeit on different points of the circuit with different bounce and perhaps (haven't tested the theory) ever so slight offset timing...at the time I was thinking they must share the burden although perhaps with an offset or proportionally.

I can settle this with an O-scope I reckon...

Anyways lets pretend I said volts instead of current....which is exactly what I actually wanted...

[EDIT]: don't need the scope, I just eyeballed the relay contacts and it's the same plate actuating both so the timing has to be parity...definitely meant volts...not current... whistlie


I just had another  facepalm moment...I can go back to MPPT. The inrush is only if I break the battery terminal...not the Solar terminal. TS PWM and MPPT incidentally have the exact same quiescent....35mA @ 50V.
« Last Edit: February 26, 2017, 06:01:14 PM by Scruff » Logged
Tinbum
Hero Member
*****
Online Online

Posts: 1177


« Reply #114 on: February 26, 2017, 07:26:10 PM »

Hi Scruff

All clear now, it was just the way it was worded, Thanks


Hi Tinbum
Quote
Hi

Quote
.I decided to series the switches to reduce the risk of arcing, switch wear and reduce the overall current per contact.

Don't understand that. Surely as the contacts are in series, the current will be the same through each contact.

If you have them in parallel the current will be reduced.

If the PV is on 9&12
GTI on 1&4
PWM on 8&5
Iain

Its done in all of my PV DC breakers by the manufactures. The more in series the higher the voltage rating.


Yes the more in series the higher the voltage rating, but he said, series to:-
Quote
reduce the overall current per contact.

depends what you are trying to reduce, voltage or current
Parallel reduces current/contact
Series reduces voltage / contact

Iain

Ah yes, I should have read it slower.
Logged

85no 58mm solar thermal tubes, 28.5Kw PV, 3 x Sunny Backup 5048, 3x Sunny Island 5048, 2795 Ah (135kWh) (c20) Rolls batteries 48v, Atmos wood gasification boiler, Brosley wood burner, 2000lt buffer tank and 250lt DHW
Scruff
Hero Member
*****
Offline Offline

Posts: 1230



« Reply #115 on: February 28, 2017, 10:47:31 PM »

I've been meaning to test that Studer 5kVA the last few days but I've been distracted by some other projects in the mean time.

Yesterday my laptop power port expired, which meant no Navitron  Shocked so obviously that took top priority!

Telling signs like this are subtly hinting it's about time to invest in some SMD rework kit.



The photosensing relay is installed and works impeccably. It took an hour or so in cloudy weather to dial it in, it cycles on and off when my panel is producing 38watts. It's impressively repeatable and accurate. I was mistaken earlier...there's no reverse polarity protection. The molex (not included) fits uni-directionally and kills the board but the leds will still work fine. Not worth fixing because the whole is cheaper than most any one of it's parts.

There is a lot of cycling on cloudy days...doesn't bother me, if it did a delay release relay could be implemented.
[EDIT] On the plus side switching the input DC side means I'm not triggering the G83/2 timer. It's straight into action.



After I broke my laptop (accidental) I decided to break (salvage) another I've had in the parts bin for a new USB port, head phone socket and tiny screws to repair my daily.

I thought the screen might be useful to remove the LCD and hotwire the led driver as another 12v lighting fixture.



I think it's art because the light output is feeble but it does make an interesting desk feature.


While I was looking for the screen backlight LED driver IC to hack I was using a head torch and a lense to search the board. The lense takes a 4.5V AAA battery but given that alkaline cells are likely the most expensive power known to mankind I rarely use them.

This thing was demanding to be hacked too.





hmmm...handy.. Cheesy



Other things of interest in the laptop were the lid hinges....no immediate use...and the battery ballspin

3 x 18650 resting at about 1.1V p/c...
I recovered it with brute force and ignorance...basically stuck it in a detonation chamber and tailed it to my 12V bus bar.  Cheesy

Worked grand!



Discharge tested 12.5V -> 9.0V: 2.0Ah

Rechargeable torch? 10W MR16 one end, voltmeter thuther!



..we can do better!



It's a Torch!



...and a USB PSU!



...and a laptop PSU!



..filled with features.

Charges direct from 12v bus (while bus is off charge (~12.6V))

Over-voltage protection: manual.
Over-temp disconnect: manual.
Over-discharge protection: manual.
Reverse polarity protection: manual.
Cell balancing: manual.
Over-current protection: fuse.

Pat pending...currently accepting commissions.  Wink

Back to the more serious business of photon wrangling in the morn...
« Last Edit: March 01, 2017, 08:49:00 AM by Scruff » Logged
jonesy
Hero Member
*****
Offline Offline

Posts: 839



« Reply #116 on: March 01, 2017, 02:11:52 PM »

Boomstick eh? Man after my own heart with all that insulation tape.
That driver chip is nice.  I came across them a month or so ago, but they're been around a few years. They make a nice compact mains to 5v psu with no transformers with virtually no idle current, so ideal for wifi projects.
The figures stamped on that relay are the ul/vde figures, RS, Farnell etc just copy the datasheet headline figures which are always pie in the sky.  Achievable, but as one very big manufacturer explained to me a long time ago, you have to do your own testing once you use it outside the vde rating.  They didn't hold up much hope for the application I had in mind for 300+ relays that I needed.
With your revised method, just keep an eye on it.  You soon see the pitting building up.If you like that ozone smell and can tolerate a shorted contact, go with it  Grin You'll be fine.
A smd rework probably won't touch a through hole power socket.  I've had to replace a few (don't ask) and the easiest method is to dremel the socket apart, then use a 50W iron to unsolder the bits left. If you have a 75W+ desolder station, that will be quicker. The pcb will be multi layer, so don't be tempted to heat and rock the part out if it have more than 2 pins; you'll just remove the through hole plating with too much force.
Stick a few diodes in series with that led magnifier.  When it was made, they over drive the leds by a fair bit.  Running at 5v may finish it off pdq. There may be a CC chip in there; a ikea lamp I hacked did.  Completely amazed me.
Where do you find the time to do this stuff?
Logged

1.1kWp PV & SB1700. 7kW log burner.
Scruff
Hero Member
*****
Offline Offline

Posts: 1230



« Reply #117 on: March 01, 2017, 03:43:19 PM »

Cheers Jonesy,

That LNK chip is deadly...even less components if you feed it 120VDC  Grin

Yeah the power port was another wrong tool for the job gig...I could do with a desolder station, a hot air station, meter needle probes, hakko tweezers, anna stoopid apprenticevolt proprietary interface...see how far I can get for now with what I have though eh...


I'm happy to torture test that little relay she's cycling aboot 20 times a day...won't be long.. bike


So my Studer 5kVA is right d#*$ered...I'm starting to notice a theme with this project...she's on the way to the bench of healing with a limping fan and an over-temp fault.  facepalm
How is every item sold untested on eblag..."swear it woz working honest, but gonna say it's broken coz I'm nice like that"...a steaming pile... fume
I don't mind broken kit, what phishes me off though are outright liars!  fume

Nah if the smoke comes outtov those lense leds (resistor) I'll just upgrade to a bodge job 12v optical ring led I've loadsa retired striplight.

Time..serious I work all day every day for a month then take a month off. Why dya think it's taken so long to get this far?

Studer autopsy reports tonight.... wackoold
Logged
Scruff
Hero Member
*****
Offline Offline

Posts: 1230



« Reply #118 on: March 02, 2017, 02:19:00 AM »

I have to say I really like this inverter. Through hole, hernia inducing, mega-surge, simple, effective!

It's a real meat and potatos build. Screw terminals on all the major power components..not this wimpy solder malarky thank you very much.

Autospy report:

This is one extremely serviceable piece of hardware.  genuflect



Mine's a right filther. I think it was residing ina metal fabricators or something, the dust inside is unreal...I had to give it a bath...it was just too much filth for me.



So I started her up with the hard power switch heard a fan grinding it's commutator and got an over-temp. fault. So that was enough for me to decommission the test rig and get her to the examination table for investigative surgery.

First things first, I needed a surgery...48V supply, meters, reversable latching 80A 80V contactor (because I hate connecting these things with the sparks and chunks of cable lugs flying off)
...oh I ought to have a current limiting facility...I was working on one but it fell to the wayside.



Contactor Coil Snubber



Prove the widow-makers disarmed. (turn on the inverter with all the power disconnected usually works to discharge 'em)



Holey FET army Batman.. faint



Hmmm toroids!



This one's had it. Two feet in the grave.



Inverter brain..better keep it safe.



Whadya make of this Jonesy?



..and R38?



Positive and negative "traces"...cwaor..no messin'!



Most of the inverter is date coded 2008. Except the FET board, I'm guessing warranty replacement in 2012 and the indicator board looks brand spanking...about the only part.



All those boards need cleaning in the morning, too late now for air compressors.  fight

Inverter bath night.



I'm not joking..I have a pretty high tolerance for filth but this thing...!



Shiney...



Good Cap.



Bad Cap.




Any ideas where I'd find a temp sensor in there then? There's a conductor set coming off the main toroid headed to a darlington transistor IC..I think it's a tap because it's v. low resistance. There's no heat sinks on this board (other than the bus, which is also a chassis...foxy Swiss!) I'll start at the fans and work back from there unless anyone knows better?

Suppose I should probably RTFM.  whistlie

« Last Edit: March 02, 2017, 02:36:31 AM by Scruff » Logged
todthedog
Global Moderator
Hero Member
*****
Offline Offline

Posts: 4235



« Reply #119 on: March 02, 2017, 07:20:43 AM »

A fantastic post, and a labour of love.  genuflect
A bit above my head.
Beautifully illustrated as well I love pictures.
Logged

'In a time of universal deceit, telling the truth is a revolutionary act'
Pages: 1 ... 4 5 6 7 [8] 9 10 11 12 ... 20   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.20 | SMF © 2013, Simple Machines
SMFAds for Free Forums
Simple Audio Video Embedder
Valid XHTML 1.0! Valid CSS!