testing the egs002 inverter board

[Antman]

Thanks Billy

Another update

Iv'e done a little bit of testing albeit limited because I only have a 200VA transformer. I was lucky enough to grab the thermal camera from work to see whats going on. Because I only had the boards made with 1 oz copper and I don't have any bus bars in place most of the current is focusing right at the center of the board.

All of the photos were taken at 26 amps of input current to the center two terminals. Once power input is spread across the 3 inputs and a solid bus bar is screwed into place on the output i'm certain the current and heat spread will be much better. I did place a tiny fan on it later and the hot spots went down to about 5 degrees above ambient.

With 2 oz copper and output bus bars in place I think 125 amps would be an acceptable limit. I'd love to do 4 oz copper but the price is 10 times more and just not worth it

[Antman]

Also got a chance to get the arduino portion up and running with a very simple program.

So far I can read the thermocouple and vary a PWM to the fan depending on how hot the unit is. The fan ramps up nicely depending on temperature which will be great to keep it nice and quiet. In the future I'd like to have the fan also react to load so the fan can ramp up before the heat saturates the toroidal.

Iv'e got it shutting down the EG8010 too through the temp feedback pin if an over temperature event is reached. This will also be used for over / under voltage and over current protection. It works very well this way because it is an instant shutdown and an instant start up. The soft start function can be used also with just a change of jumper on the EGS002.

The next revision i'll have to put in an extra thermocouple so there can be one for the board and one for the transformer.

This weekend I'll work on the programming but i'm more of a hardware person so we'll see how that goes

Made a short video to show it off...

[oztules]

nice to see it running Antman

will be interesting how much current you can push through it safely in the finish.

125amps is plenty for 48v use..... be a very good outcome if you can achieve this easily.

............oztules

[frwainscott]

Antman,

Impressive... Did you cut the  393 chip off of the 002 board or is it still working?

Also... I think that's the way to go with the board,fets & heatsink. Like the one Oz had showed running, makes it possible to mount in A multitude of scenarios. Makes for a much cleaner package.

Cudos to both you and Oz.... :-)

Frank.

[Antman]

Thanks Frank and OZ

I think I like the layout but I just have to confirm it with testing to make sure it's solid.

The 393 chip is still in place and the EGS002 remains unchanged. The current feedback pin which is the problem is grounded in my design so is taken out of the equation all together.

[billy]

Oztules,
 just wondering if it would take much to change a 24 volt pj controll card to 48 volts, or should I just make one of yours.

Thanks,
Billy

[oztules]

The pj card is complex, and I don't know what the real differences are.... maybe just make a 002 or 8010 device , it works every bit as well... probably cooler from what I have found.

..............oztules

[billy]

ok Oztules,

thanks again, I'll make your latest control card.

oh by the way great work your doing.

Billy

[ClockmanFrance]

Hi Oztules,

Its a stupid question from me, .....

Have you done a pin for pin connection from your latest 8010 control Board to the PJ Board.?

Thanks.

[oztules]

Yes, pin compatible.
How they coming along?

.......oztules

[ClockmanFrance]

Thank you very much.

Yes both boards done, here's the Power Board.......

Putting it all together now.... for the BigOzInverter, then we will see....... gulp!.

I do not like spade connectors on the control board at 10mm/2, but another insulated box with the CT sense coil and  two bolt and eye connections seems a bit OTT. So thinking about the secondary output cabling if I can feed back to the control card at my EMC Filter assembly unit, where the two Secondary cable would have its first connections, ? I would use 10mm/2 cable. Could put sense coil on the secondary before the EMC filter.

The control board is just single sided with a few top jumpers. Those SMD connections need a steady iron, by the 3rd chip/ card I had it sorted.

Just got a batch of 30off  8010's with the HT1536.

[ClockmanFrance]

...and

Note.... the present PJ board has the long 48v + heat sink connecting directly to the underside copper. I have left mine as was using the 12 FET centre legs. ?

[oztules]

The 12 fet legs will do the job, but no crime in using the 4 mounting bolts as well....may take some of the pressure off the legs.

No current need go to the control card now we use the CT coil. leads can be as long as you need to get to where the power leads are.... we need only 240v for correction, and the feedback from the CT coil at the control card.. spade connectors are way over the top in this case, but convenient.

Yes, the first few 8010 were best not seen by the populace, but after that it became simple.

Do you have the control card as yet ( pic?)

Main boards are better than the original by a wide margin... I always hated the close proximity of the lands with the fet legs.. and all other inter track spacings for that matter.... all too cosy for my liking.... at least you have decent spacing.
Also like the way you have spread the pin area with solder to get more power over a wider area before getting lost in the vast area copper.

............oztules

[ClockmanFrance]

Thanks for reassurance on the AC side.

I will amend the Power boards for that extra connection on the base of the 48v + heat sink, as you say it won't hurt.

The OzControl board, pretty much like yours, but here I have altered the size to take suitable mounting positions on the Power Board Heat sinks.

 I have beefed up all copper runs, because I will also supply the masks for Home made PCB's and drilling holes with the aid of a template gives a bit of leeway for the skill of the driller.
And your phrase, "there was a fracture in one of the copper thin trace" or something similar?

[ClockmanFrance]

and.....

Again its keeping things real simple and yet robust, so folk anywhere in the World can follow and get a result.

The drill mask is also helpful in drilling the Heat Sinks. The cable mounts are threaded 6mm dia, with all other mounting holes at threaded 3mm dia.

The FET legs are 18mm from PCB to mounting hole, so gentle bends are easy and the legs not to long.

Rubber strips, from 1.05mm thick from a rubber Pond Liner offcuts, for the heat sinks insulation.