I have a TL866CS modded to TL866A, and was able to read but unable to write ST M27C256B, and AMD Am27C010 UV EPROMs.
The AMD chip was from a Socket 3 486 motherboard, I was able to read the contents of the ROM BIOS, and use the resulting image to boot the PCEM emulator. The ST chip was of unknown origin, but the resulting image contained a handful of human-readable text throughout, so I can assume it dumped OK.
I erased the chips in a cheap eBay UV eraser for 20 minutes, and verified that the chips read out all FF's. They both passed this test.
When I attempted to program these chips, the problems began. I decided to write all 00's as a test. The AMD chip fails on the first byte: 0x0000 Buffer=00 Verify=FF. It's clearly failing to write anything at all.
The ST chips fail in more of an insidious way: The programming process will get to some random stage and fail with a similar error: 0x**** Buffer=00 Verify=C6, where C6 is some random value. On reading back the contents of the ST chip, garbage is displayed, not even matching up with the alleged "Verify" value from the failure message.
I suspect that Vpp is not getting high enough, but my oscilloscope is still packed away from moving house, so I can't verify that just yet.
However, say it is, has anyone else experienced this? And if so, were you able to resolve it by providing additional 5v power externally? That's something I can try easily with what I have to hand.