Studies for a driver for the PMT HV cascade in CHAOS

Design guidelines

• The output shall be a sinus function.
• At the output shall be a capacitor of at least C₂=100pF, that may be mostly parasitics.
• The output amplitude will be close to 100V.
• The transformer will provide some amplification ratio.
• The current in the output loop will be thus amplified on the primary side.
• We do not want that current to be driven from the power supply directly, so there needs to be some resonance on the primary side as well.
• f=125 kHz seems to fit quite well with reachable inductances and reasonable current scales.
• A single asymmetric drive may be sufficient.

After a serious math excersize this circuit is now understood, hopefully.

Formulas

• a = Uout/U₁
• b = n₂/n₁
• c = n₃/n₁
• ω = 2πf
• C₁ = C₂ a (c²/(a-b) - b)
• L₁ = (1 - b/a) / (c²ω²C₂)
• n₁ = √(L₁/Al)

Some Configurations

This core does not work at this frequency. This has been tested on the bench.

• core B64290L0038X038
• f = 125kHz
• Al = 4.07µH
• a=5, n₁=10, n₂=30, n₃=40
• C₂=100pF, C₁=2.5nF, L₁=409µH

Faster bigger ferrite, N30:

• core B64290L0045X830
• f = 125kHz
• Al = 2.77µH
• a=7, n₁=10, n₂=40, n₃=50
• C₂=100pF, C₁=3.03nF, L₁=278µH

Bigger C₂:

• core B64290L0045X830
• f = 125kHz
• Al = 2.77µH
• a=7, n₁=6, n₂=24, n₃=30
• C₂=270pF, C₁=8.2nF, L₁=103µH

The slow ferrite at 25kHz. This will be tested with the transformers already made, see above.

• core B64290L0038X038
• f = 25kHz
• Al = 4.07µH
• a=4.65, n₁=10, n₂=30, n₃=40
• C₂=2.2nF, C₁=68nF, L₁=409µH