ADP5070 has no negative voltage

I'm having the same issues in regards to my design.

I cannot for the life of me get the negative voltage to do anything, even in LTspice.

My design roughly follows the Designer Spreadsheet for a 27V/-9V design.

in LTspice:

The simulation does absolutely nothing and ends immediately when EN2 is high:

EN1 high on the positive side everything works as expected and I'm getting 27V output, but the negative side remains dead.

Here is the LTspice file:

https://gist.githubusercontent.com/AndreiLux/8bf7217143ba725963a0ce6495fe0e50/raw/dcca2867597610c049e20d7d1430142611fd28ec/ADP5070_test.asc

Been trying to debug this for a few hours and can't figure out what's happening.

This ended up being my own stupidity in LTspice as simulating it without startup from 0V doesn't work correctly. I also had a typo in the negative side resistor at 53.6Ohm instead of 53.6kOhm.

Hello AndreiF, well finally the fault was easy. Now my design works. The goal has nothing to do with simulation or other "ideas" behind any tools. The solution for me was, that I did not resprect the given table on page 23 showing the needed values for RFT2, RFB2, Cc2 and Rc2. That was all. The COMP2 is little tricky here. Obviously. My design is running at 1.2MHz, this is the upper entry dealing with correct values in the table. 15k and 68nF are NOT good and are not the recommended values. In my design I now use 1.62M, 133k, 47n and 8.2k and VOILA, now I have exactly 9V (-9V) and all works perfect now. Of course table 11 has to use for the negative inverter. Yes, EN2 must be enabled to get the second switch working. Cvref 1uF was also changed up to 2.2uF with not seen effect, so 1uF is still ok here. For the upper switch I also changed all values to the given values on table 10. And 9V works perfectly here, too. The COMP1 accepts more variances in values than the lower switch... Cheers, good luck, Thomas

Hello AndreiF, well finally the fault was easy. Now my design works. The goal has nothing to do with simulation or other "ideas" behind any tools. The solution for me was, that I did not resprect the given table on page 23 showing the needed values for RFT2, RFB2, Cc2 and Rc2. That was all. The COMP2 is little tricky here. Obviously. My design is running at 1.2MHz, this is the upper entry dealing with correct values in the table. 15k and 68nF are NOT good and are not the recommended values. In my design I now use 1.62M, 133k, 47n and 8.2k and VOILA, now I have exactly 9V (-9V) and all works perfect now. Of course table 11 has to use for the negative inverter. Yes, EN2 must be enabled to get the second switch working. Cvref 1uF was also changed up to 2.2uF with not seen effect, so 1uF is still ok here. For the upper switch I also changed all values to the given values on table 10. And 9V works perfectly here, too. The COMP1 accepts more variances in values than the lower switch... Cheers, good luck, Thomas