Thread Rating:
  • 2 Vote(s) - 5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
48V 5000W Oil cooled inverter build
#61
More nice work :-)
How are the 23kHz traces looking at the fets? Are you seeing ringing there? Might be a source of higher losses, eg cross conduction pulses?
Are the traces different on the fet side with the T400 ferrite vs the side direct to the transformer?

The hum you're chasing, could this be "magneto-stricting" of the core, eg at peak of 50Hz current it "tightens" & when current = min, it "relaxes", result = hum?
Wondering if supporting the transformer on the edge per the pic leads to hum coupled to timber box?
If transformer was supported at centre would that be better?
Running off solar, DIY & electronics fan :-)
Reply
#62
Thanks, I had thought about hooking the scope up and was a bit lazy / tired and just plugged the wind grid inverter in as one screen shows a waveform output. At that point I was just interested in seeing the right shape and no shape horrors. When I get around to it I will hook up the scope again because I am now a bit more curious. It's nice and relatively silent now though,

The hum can also be the build up of the flux in the transformer due to resonance as the L(winding)/C(output cap) approaches 50Hz, which eventually pushes to saturation and the excess/additional field strength starts to push/pull on the windings more, so if any have room to move they will and then hum.

After researching magnetics for a few weeks for the build I'm quite amazed at the dual magic of a magnetic field. What amazes me most is a coil can take a bit of energy just to create the magnetic field and then that very same field can be reversed and thrown around at 50Hz with near zero additional energy input. That is what seems like a magic trick in some ways.

I did think I should just dump it into a tub of oil and then up-rate it to 1500W. When the battery has a bit more charge and the input voltage is around 52V I will see how the 2kW fan heater performs in terms of sag on the output voltage as it may have been partly due to the low input voltage on the battery pack at the time, or just due to the 200% loading of the transformer, or the different winding ratio.
If you can't quantify how much they cost, it's a deal, I'll buy 5 of them for 3 lumps of rocking horse ......
Reply
#63
(10-01-2019, 09:10 AM)completelycharged Wrote: The hum can also be the build up of the flux in the transformer due to resonance as the L(winding)/C(output cap) approaches 50Hz, which eventually pushes to saturation and the excess/additional field strength starts to push/pull on the windings more, so if any have room to move they will and then hum.

Well there's your problem!! You should be running it at 60Hz! Crazy EU power systems trying to be different Tongue
completelycharged likes this post
Proceed with caution. Knowledge is Power! Literally! Cool 
Knowledge is Power; Absolute Knowledge is Absolutely Shocking!
Certified 18650 Cell Reclamation Technician

Please come join in general chit-chat and randomness at Discord Chat (channels: general, 3d-printing, linux&coding, 18650, humor, ...)
(this chat is not directly affiliated with SecondLifeStorage; VALID email req'd)
Reply
#64
Yeah, oscillating low loss coil+cap circuits are weird. Exactly what a magnetic field is, is fun too (know what it does, but not what it is!).
It's almost like you need the gap an E core transformer has for a bit of loss - ie what a torroidal one basically doesn't have.

Maybe some of the sag under the 2kW load might be resistive losses in the primary side wiring? Wire doesn't look that thick for ~40A (average) operation?
Bit hard to tell your actual wire sizes from the pics!

I'm a 50Hz'er too - each time I've visited the USA, I really noticed the 60Hz buzz - funny what the brain "tunes out" after a while ;-)
Running off solar, DIY & electronics fan :-)
Reply
#65
New board arrived and finally had chance to have a look at it, which initially looked really good with additional heavy solder and wiring on the back for higher amp handling and four terminals for the -ve supply, which was a bit of a concern with the previous board only having 2 for current distribution. Was all looking really good, until I noticed the buffer capacitors are the wrong voltage.

(edit) Story ongoing... interrupted by brief power outage to the area (big enough to show up in the grid demand at around 500MW).

Adding another very very basic aspect, tollerance.

63V rated capacitors with a +/-20% tollerance rating will vary in actual ability from 50.4V to 75.6V so for those that are more than 11% (56V) out of tollerance on the low side for me they will fail and even then limit me to using the board with any other higher voltage.

For the sellers own information, who states the over voltage protection limit works out to be 64V for the 48V board this is already above the capacitor rating even before factoring in tollerance.

More on the capacitors : http://www.thebackshed.com/forum/ViewTopic.php?TID=9085

Just really, really, f*** annoying they can't just say, ah sorry, sent you the wrong board and instead come out with absolute b*** s**t "The maximum voltage of 63V capacitor can reach 85V" I kid you not.

Based on the 63V 85V comment this could imply he thinks they are +/-35% in which case the lower actual possible is 41V.... bang !!!!!! I have a 36V board at best.
If you can't quantify how much they cost, it's a deal, I'll buy 5 of them for 3 lumps of rocking horse ......
Reply
#66
(10-20-2019, 06:48 PM)completelycharged Wrote: (edit) Story ongoing... interrupted by brief power outage to the area.

Hahah, ironic Tongue
completelycharged likes this post
Proceed with caution. Knowledge is Power! Literally! Cool 
Knowledge is Power; Absolute Knowledge is Absolutely Shocking!
Certified 18650 Cell Reclamation Technician

Please come join in general chit-chat and randomness at Discord Chat (channels: general, 3d-printing, linux&coding, 18650, humor, ...)
(this chat is not directly affiliated with SecondLifeStorage; VALID email req'd)
Reply
#67
So, so, so ironic. Managed to hot wire the 1kVA unit up after finding some wire (had packed the unit for a trip) hooked it up, switched on, 600W, house powered, just...... few minutes later power comes back on.... doh !!!!
Korishan likes this post
If you can't quantify how much they cost, it's a deal, I'll buy 5 of them for 3 lumps of rocking horse ......
Reply
#68
ROFL  double irony!!  Tongue Big Grin
Proceed with caution. Knowledge is Power! Literally! Cool 
Knowledge is Power; Absolute Knowledge is Absolutely Shocking!
Certified 18650 Cell Reclamation Technician

Please come join in general chit-chat and randomness at Discord Chat (channels: general, 3d-printing, linux&coding, 18650, humor, ...)
(this chat is not directly affiliated with SecondLifeStorage; VALID email req'd)
Reply
#69
So, after a diplomatic excersize and a climb down off my high rocking horse and many communications between the seller and myself he agreed to send me out some 100V capacitors. Will have to unpack the board again.

Anyhow, hopefully when I get some 100V caps I will then swap them out and power up the board.

The ferrite cores that I wound and had been previously using are currently having a large dose of superglue to fix them together so that they do not vibrate and reduce or hopefully eliminate the noise. Extreme option would be to put them in epoxy to make them silent but they would likely overheat OR...... in an oil bath.

Current plan is to use an old computer case to put the board and transformer in, however the weight of the transformer is a real issue as most computer cases are built for 1kg electronics boards and not 20+kg of copper and iron.

The transofrmer at the moment will sit (not fixed) inside the case while I have a think about a fixing method that will work when it is dropped, lol.

The case is a cube case and split so the other half is then where the inverter board, filters, capaictor and chokes will go if I can squeeze them all in.

The chokes need a bit of consideration as to where and how they are fixed as the ferrite cores will otherwise treat part of the fcase as a nice magnetic loop and alter the unductance. For this reason fixing them directly against the steel without a spacer is not an option.

Will hopefully have the board and parts arranged in the next day or two and then post a picture,
If you can't quantify how much they cost, it's a deal, I'll buy 5 of them for 3 lumps of rocking horse ......
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)