Guru
Joined: 21/06/2018
Location: AustraliaPosts: 1173
| Posted: 02:42pm 25 Jun 2024 |
NOTE 2 Changes to WG39Rev7 one an omission on the BOM (Q11) two a component value tweak.
If you already soldered in R7 sorry it should be changed from 150K to 120K see below.
I have just finished building the inverter circuit board WG38Rev7 which was recently shipped out. Rev7 had a few extra parts added to enable accurate setting of low AC currents ie 0.2A as well as ability to set the AC zero voltage. Ok I admit this was overkill so by all means leave out the trimpot RV4 - it is unlikely you want accuracy of your inverter near zero volts out (leaving out R4 will not cause any other issues). Using Poidas code, when the inverter is not producing any AC output the error messages cover the VAC value anyway but possibly the KeepIS code might do this differently and ACV is visible albeit with zero volts so if there is a residual reading it should be able to be offset back to zero.
I found 1 error using my BOM, the line Q3, 4, 6, 10 - should read Q3, 4, 6, 10, 11 and the quantity should be 5 not 4. Q11 was added in the last revision it allows the indicator for Vcap/Vlow to use either 5V or 12V to drive the LEDs. The front panel is currently joined with a small track to power it from 5V by default. If you have 12 V LEDs and want to use them then the track needs to be cut and a link inserted to power the front panel from 12V.
Right next to the J3 silkscreen designator, you will see 3 pads, the outer two are labelled +5V and +12V if you look closely you will see a track on the component side of the PCB joining the centre pad to the +5V pad, cutting this track and then linking the centre with +12V pad will give you 12V on the front panel. One word of caution the current being sunk by the Q11 should not exceed ~30mA max as the current is limited by the current capability of the Nanos D13 output pin, so it needs to be a LED not a lamp.
After building up the PCB and testing, it worked first go BUT there needs to be a component value change R7 is listed as 150K, please use a 120K as the linearity of the ACV above 230VAC started to suffer (a lower reading). Although it was not noticeable on my previous test unit (@235V) it needed to be scaled down a bit and is now linear to over 250VAC.
Forgot to mention I installed the 4 pin linear regulator with its zener and resistor and it works great, at 11V in still working fine at 10V in has shut down the 5V cleanly.
Poida is tidying up the code for a proper release version in due course, if anyone needs code before that release there is a version available that works ok but the newer release will the best one to use.
Mab your board arrived in UK ~24 hours ago so its getting closer, Ziki & Nick your packages are awaiting the next flight from Melbourne - I dont know why they were a bit slower, they were all posted at once....
Edited 2024-06-26 00:45 by wiseguy