Tito

PS RE your sample-escaping problem, this is where vertical tubes come in handy – in my case I drilled a bunch of holes in the bottom of a plastic box for the upper chamber (P1000 tip box lid), wedged some small-diameter rubber tubing in the holes, wedged the pipettes with gel in into the tubing, and then suspended it over the lower chamber so the bottom ends of the gels hung free into the buffer. Another advantage of this layout is that the current is only flowing through the gel, not through the whole cross section of the buffer box, so the total load on your power supply and resistive heating are minimized – IIRC (it was >10 years ago), this is why I did it that way. I used a conventional bio lab voltage source for it though, can’t remember the settings but >>9V for sure – might have even been a thousand.

PPS RE Run #2 – from your image I am going to guess that if you had linear electrodes across the full width of the gel box at both ends, both straws would have worked. I think maybe the use of “point” electrodes is giving you a funny-shaped field that may also be reducing the quality of the separations in the straws that do work (although the nice gel image on the left may argue against this, so maybe I’m wrong).

The slightly alkaline TBE and TAE buffers (which aren’t exactly exotic) usually used for nucleic acid electrophoresis tend to keep the DNA/RNA deprotonated (-) to allow for better resolution. An unbuffered salt solution may not resolve much in terms of DNA.

And another note on standardization. Another reason slab gels are popular is that they allow you to run a molecular weight ladder along side your sample, adding additional assurance that the DNA you desire is the correct size and allowing comparison between separate gels.