Tomm,
Very interesting information here. I am wondering if any of the
commerical tube amps are using this kind of circuit for the plate voltage.
Given a decent 2KW TX alone can set easily set a guy back $400 or so, a
switching power supply seems the natural evolution for tube amps. But
I've not seen a switcher in a tube amp yet.
Is there is some limitation that prevents easy scaling to those power
levels and voltages?
73/jeff/ac0c
KD7QAE wrote:
> Paul,
>
> At a 6:1 ratio, assuming the core are supports the flux created by 340V
> and 6 turns, you will have an output of 6x340 or 2040V. The transformer
> takes the 680Vp-p primary square wave and delivers 4080Vp-p to the FWB
> which 'inverts' the negative pulses and adds them into the gaps between
> the positive pulses to give you the 2040Vdc output.
>
> Since this is a square wave, a capacitive filter will suffice to fill in
> the small gaps due to switch rise and fall times and transformer
> response. I would build this filter as a high frequency filter followed
> by a few hundred uF of energy storage caps.
>
> If you want to pull the maximum out of the AC line you have to power
> factor correct the input rectifier so the PS looks resistive to the
> incoming AC. This is a simple matter of either building a PFC circuit
> from PS controller application guides, or, better yet, buying a surplus
> 2kW computer (server) power supply and reusing the AC front end to get
> the PFC and even the power MOSFETS. By the way, this approach will give
> you line regulation and, if you reuse more of the PSU, load regulation
> as well.
>
> On a final note, while the 50% square wave chopper is an easy circuit, I
> see a major drawback in that it has no load or line regulation. If you
> were to add a simple PWM controller to this so that the duty cycle of
> the square wave were variable, and add an output inductor filter, you
> would then have a regulated HV PSU for not much more trouble than what
> you are now building.
>
> Tomm, KD7QAE
>
> Paul Decker wrote:
>
>> I've been holding this question for a couple of days now, I'm sure it is
>> very simple and perhaps I just need some reassurance on the answer.
>>
>>
>>
>> If you have been following some of this smps discussion, I've got 100Khz
>> pulsed DC (0 - 340v) which is generated by directly rectifing and filtering
>> the 240 V AC mains and providing that into an h-bridge. The h-bridge dumps
>> the 340 V 100Khz square wave into the transformer.
>>
>>
>>
>> As the QST article recomends, I've wound the transformer with five turns on
>> the primary and had calculated that I need 30 turns on the secondary.
>> Performing some small signal tests, inputting 3.4v pk-pk square wave from my
>> signal generator yeilds about 20.4 volts pk-pk square wave. I believe this
>> relationship should be linear and inputting 340 V will yeild 2040 volts on
>> the secondary.
>>
>>
>>
>> At this point the secondary dumps into a full wave bridge rectifier followed
>> by a filter capacitor. This is where I am unclear. When I rectify this
>> with the full wave bridge, will I get 2040 * 0.90 or will I get 2040 *
>> 1.414 as the final DC output? Part of me says I get the 1.414 value of
>> 2885 VDC, however reading through the handbook, I seem to be reading I'll
>> get 0.90 the output voltage.
>>
>>
>>
>> thanks,
>>
>> Paul
>>
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> Amps mailing list
>> Amps@contesting.com
>> http://lists.contesting.com/mailman/listinfo/amps
>>
>>
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|