Already for quite some years I was not happy with my Begali / microKeyer II combo. Sometimes the paddle action just does not yield the right output from the microKeyer. The microKeyer employs the WinKeyer v.1. Back in 2006 there were already reports that this chip does not deal very well with certain contact alloys (e.g. aluminium) due to oxide layer build-up on the contact surface (re-iterated by Joe Subich in 2021, here). Begali has changed their contact material after that. Now, I ordered my Begali around 2005, so I might have the initial contact material, explaining this issue. A debouncer was made available by microHam which was supposed to cure this issue, however, after using it for a while I concluded it did not solve the issue entirely.
It was only until last week when my friend Jo, PG4I, pointed me to an article by a Danish amateur, OZ1JHM, describing the so-called minimum “wetting current” of an electric contact. This seems to be applicable to the issue at hand !! So, I measured the current through the paddle contacts and it appeared to be only about 0.5 mA. That is definitely not enough for a reliable contact action according to the article. If you look at the schematics of the WinKeyer 1 (which is undoubtedly copied by microHam), you see that the paddle input circuitry comprises a high impedance voltage divider (R6, R9, R10) to establish the right voltage to the Paddle Input pin of the WinKeyer chip (see specs, p.4). Rather than changing this to a low impedance voltage divider (divide all R6/9/10 resistor values by 20) to increase the current through the paddle contacts, you can simply add a lower value pull-up resistor in parallel to the voltage divider. So I soldered two 680 ohm resistors from each paddle contact input to the 5V supply rail. Pictures below explain it all …
There is now about 8 mA flowing through the paddle contacts, which seems to be enough. Problem solved ! I am now a happy camper and keying smoothly on my Begali Signature Ed.
BTW. Just read on the groups.io forum that the microKeyer II has a build-in debouncer already. This just shows that bouncing of paddle contacts is not the actual issue. With the debouncing circuitry the current through the paddle contacts is still too low (0.5 mA, see above) and you need the pull-up resistors to increase the contact current.
BTW 2. I tried to post this improvement on the microHAM groups.io forum, but my posting was not approved. The reason given was: “Unauthorized modifications to microHAM product”. I do not know who the moderator is, but I assume it is Joe, W4TV, who apparently is afraid to loose his face … The uKeyer II is out of production and therefore all existing ones are out of warranty for a long time. So, why a super simple improvement like this can be an “unauthorized modification” keeps me puzzled. It is certainly not in the spirit of ham radio. Shame on the moderator of the microHAM groups.io !!