CH341A Serial Memory Programmer Power Supply Fix

[amyk]

The 5V / 3.3V myth perpetuated by the internet was evaluated and debunked a few years ago.  No need to mod the CH341A.  Just use it . . .

Have a look at this video created by Sorin (Electronics Repair School) analyzing the issue in detail:  https://youtu.be/J8-Sh7DjiXw

Given that these are most commonly sold as BIOS programmers with clip leads for in-circuit programming, I wonder if the decision to put 5V on the data pins and 3.3V on the power was a "hack" to make it work in-circuit; I looked at a few random laptop schematics and some have a diode supplying the Vcc to the EEPROM while the other I/Os are connected to the rest of the circuitry, often via a few dozen ohms of resistance. With that arrangement it makes sense to have Vcc at 3.3V, but maybe supplying 3.3V on the data pins (via the output impedance of the CH341) causes the rest of the circuitry to act as a voltage divider, pulling down the voltage below Vih(min), and they figured that making it output 5V unloaded would work?

Given that context, perhaps using the programmer on a mobo in-circuit won't cause any problems with the 5V unloaded outputs (check the schematic to be sure), but programming removed chips directly is a bad idea?

[Moij2x]

I have question about CH341B programmer (3.3V from production).
Is possible with this programmer read (later maybe write) EN25Q40 SPI memory without soldering chip out from PCB?
PCB is Samsung TV board bn91-21809k made by Hansol.
I tested it when PCB was without power. I use Neoprogrammer software, installed WUDF driver, programmer was recognized by PC as USB-EPP IC CH341A. All seems OK also Neoprogrammer software recognize programmer, but when I connect adapter to EN25Q40 then PC and also Neoprogrammer software lost CH341B programmer connection. Supply voltage on EN25Q40 when CH341B was connected, was 3.2 V.

I forgot to add that I am 99% sure programmer was correctly connected to chip, both ends where double checked.

[Moij2x]

I find some information on other forums that problem can be solved adding capacitor and resistor to 3.3V regulator output, voltage here are not very stable. I will do it later.
AMS1117 datasheet have information "The addition of 22μF solid tantalum on the output will ensure stability for all operating conditions" This capacitor is missing, there is very small, about 40 nF capacitor.

[Moij2x]

20 uF capasitor and 280 ohm resistor from 3.3V to ground and 100 uF from 5V to ground solved the problem.