Look at scrap yard collecting electric/electronic waste. They may sell you dismantled heatsinks fro old electric apparatus. Alternatively old TV set contain an aluminium heatsink but not very pretty looking 

hi forest,

If you noticed carefully in the old tv set the old heatsink design normally doesn't comes with fins to dissipate heat properly although it tend to appear larger like a cut aluminum plate.
Only sound amplifier heatsink tend to be larger which is harder to spot in scrap yard.

Nick,

You may have the two halves your ferrite trafo primary in anti-phase. A photo of the underside of the primary winding will enable me to determine this.

  Hoppy: I've already reversed the secondary coil of the flyback core, and didn't see much if any difference. Is that what you mean by anti-phase? 
  I've also added (in series) the white LC coil wire to the thinner red output coil. It raised the voltage, but not much difference in bulb brightness. Nor can a 100w bulb be lit, yet.

  Havuhung:
  That is an important question.  Unless it runs on its own power,  probably not.

   Hoppy: You've mentioned that your RMG replication can light several 60 watt bulbs, but you have not shown that, it would be good to do so, when you have a chance.

  Geo has also mentioned that his device can operation up to 3000 watts output, but again, he has not shown that, as well, only partially lighting about 300watts.
 
  So, how 'bout if we can agree to show what is being mentioned, and can be verified by a video, or working example, etz... As this can only bring doubts. Which it seams that there are already plenty off.
 
  I have not figured out what may be wrong with my set up, yet. As it will not light even a single 100w bulb, like Hoppy's crt couldn't, also.  Much less the 460 watts worth of bulbs, which the same Mazilli crt connected to my yoke core can.
  Seams to me that the flyback core, as well as the air cores with the ferrite pieces are not very efficient, or as good of an induction coil set up, as the bigger ferrite yoke. Although my yoke core does draw lots of amps,  this Mazilli/Geo coils replication runs totally cool, at least so far. That's one good thing.

  I'll be connecting up the Mazilli crt to the yoke core, again, and also to the previously made Akula air coils. The idea is to try to see if I can replicate the original Akula 750 watt device, using this Mazilli oscilator/yoke core, air coils,  etz...

  I hope that Geo joins us in further tests, meter readings and such, and also showing his set up running the 3000 watts that he mentioned, while metering the results.
  We could use some help, as Itsu, Hoppy, myself, and several others aren't getting the
optimum results, yet.

   a.king:
   I wish that I were getting excited. But, after spending several days to replicate what I'm working on, no, I'm not very excited yet.
  Much less with the Akula devices that no one can replicate, or the no results from all of the TK replications, or ANY other self runner that has being tried to date, by US.
  What is it that we should be getting excited about?   

  Hoppy:  My comments were pointing towards the lighting of ANY set of several, (not one), 100 watt bulbs.  Optimum results is opinion based. But, showing the RMG lighting 3000 watts worth of either 110v or 220v bulbs, or halogen or ANY type of higher output bulbs is what I was getting to, as being optimum. And showing it self-running, if only for one minute, would be like icing on the cake. But, no such luck.