[Home][2017 Index] [TCML] Fw: Ballast Inductor [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[TCML] Fw: Ballast Inductor

David, I would love some pictures of your ballast. =) 
I think I have your old plastic capacitor btw.   I can't remember who I bought the capacitor from, but I am pretty sure it was you. It's still going strong.
I don't think I have seen #8 wire for sale at Lowe's. Definitely #6 though, I'll have to check for #8 the next time I'm there.  But to be clear, you are sourcing between 40 and 50 amps through a single #8 wire and with your airflow, it seems fine?   With 2 parallel coils for 80-100 amp total current?
I took some pictures/measurments and uploaded them to 
Precise measurments say that the total gap is 1.2" long, distributed across 12 gaps.
I put a dimensioned drawing in the link.
I wrapped some thin-ish wire around the core, and measured 15mH or so with ~125 turns. That dropped to ~6mH when I introduced 2 inches of wood into the magnetic path. That was at ~1A. 
The inductor was rated for ~130amps, before I removed the coils (inductance was around 120uH, the name plate I have in the link is for the larger core I think)  and came from a high power 3 phase ~2001 time frame uninterruptible power supply. I am not sure what output voltage it was intended for but I get the impression it was more than 120V.   The 3 bridge caps were rated for 460V. Perhaps a core with such a large gap won't saturate at the currents I desire. 
I have 2 medium sized cores, of which i dissassembled and dimensioned one of them. I have 4, that are slightly smaller.And 1 that is slightly bigger. 
If I combine my 2 medium cores I'll have a cross sectional area of 9.3in^2 give or take. If I combine the 4 small cores. I'll have >16in^2, probably...
But all these cores have significant gapping built in already. 
Is there a rule of thumb for how much every mm of gap increases the saturation point?
Or how the inductance would vary as more current is applied?
Or is expermental analysis the only way guestimate such things?I don't have a way of sourcing 100A at the moment, but designing for that eventuality makes sense. If I can get away with using just 2 cores, that will save on weight and wire which would be nice, but using all 4. 
Thanks, -Jay
------ Original message------From: David Rieben Date: Fri, Feb 17, 2017 8:03 AMTo: jhowson4@xxxxxxxxxxx;Cc: Subject:Ballast Inductor
Hi Jay,I am unaware of the physical dimensions of the core that you have acquired, but I personally salvaged a core from a defunct old style x-ray transformer and constructed a homemade ballast that easily handles 80 to 100 amps as much as I need to push that much power. I can't recall the exact dimensions of the core off the top of my head, but the total weight of the core assembly was a good 90 to 100 lbs. The core was in two pieces, each piece in the shape of a "U" and the laminations were well glued together into a homogeneous mass, so no concerns for laminations coming apart. Each of the core "U's" resembled a large, old style horseshoe magnet and the longer "U" was the half that the original windings were installed. The shorter "U" was the 'cap' that completed the magnetic circuit loop.I simply acquired a few hundred feet of #8 THNN stranded and insulated building wire from my neighborhood Home Depot (or Lowes?) and wound about 105 turns on each leg of the longer "U" and electrically paralleled the two coils as a single ballast coil. I determined the proper spacing between the longer "U" core with my newly installed windings on it and the shorter "U" to complete the magnetic circuit loop by trial and error and set the spacing between the two "U's" so that the ballast coil would draw about 100 amps when shorted across a 240 bolt AC source. Seems like the optimal core half spacing ended up between 1/4" and 3/8". The greater the spacing, the greater the current draw.As a side note, I spaced the individual layers of the two coils form their neighbor layers by about 1/8" to facilitate enhanced air circulation and cooling and then I finally installed a small fan to force cooling air through the coil layers during operation.This setup has performed reliably and given me solid and worry free performance inside the control panel of my Green Monster coil and also handily drives high powered resonant plasma arc generation or 'Jacob's Ladders'. Keep in mind that like with most coilers, my use is going to be intermittent and of relatively low duty cycle, but I have never really been able to detect any notable warming of the coils of my homemade ballast above ambient temperature after operation.Just let me know if you would like further details or pictures.David Rieben Sent from my iPhone
Tesla mailing list