[Collins] 516-F2 PS Question - some numbers

[Dr. Gerald N. Johnson]

When the scope peak to peak and the alleged RMS have a fairly consistent 
2.8 ratio, I'll allege its peak to peak divided by 2.8. The peak to peak 
reading is falling off a log simple to make in a meter, while really 
true RMS is very difficult to accomplish. There are some approximations 
that work for some waveforms, and fail for waveforms like that of 
rectifier ripple.

73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association.

On 11/11/2015 4:09 PM, Paul Kraemer wrote:
> I don't think you can make that declaration if the Fluke happens to be a
> true rms reading device which many are.
> Paul
>
>
> -----Original Message----- From: Dr. Gerald N. Johnson
> Sent: Wednesday, November 11, 2015 4:00 PM
> To: collins at mailman.qth.net
> Subject: Re: [Collins] 516-F2 PS Question - some numbers
>
> The fluke's RMS reading is almost sure to be based on a peak to peak
> detector with the value divided by 2.8. Its not true RMS because the
> waveform is not sinusoidal when its ripple. Its essentially the sine
> wave squared or full wave rectified. Has lots of harmonics.
>
> My question is what was the load on the supply? The ripple and the
> effect of the choke tuning capacitor are significantly affected by the
> load, and the choke inductance usually is sensitive to the DC load current.
>
> The ten volt rise in the last test could easily be line voltage change
> from a load in the area being turned off. That's only a 1.17% change.
> Well within normal line voltage variation.
>
> I presume that the original designers tested the circuit with the tuned
> choke under varying load and varying line voltage. I'd be inclined to
> stick with the value they chose.
>
> 73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association.
>
>
>
> On 11/11/2015 11:01 AM, Al Parker wrote:
>> Hi Folks,
>> Well, after reading all the replies, and checking my inventory, I
>> decided some numbers might be interesting. So I took measurements using
>> the Tek 7xxx scope and Fluke meter, with different capacitor values
>> across the filter choke in the HV secn of my "bench" 516F-2 that's the
>> subject of this whole exercise. I had replaced the electrolytics and the
>> bleeders in this supply some mo.s ago, maybe over a yr ago now.
>> I read the peak to peak 120 cycle ripple value on the scope, not super
>> accurately, and took the AC rms reading with the Fluke. The values match
>> roughly with the math. HV read with the Fluke.
>> I set the line input to give 6.3 vac on the filament line, and that
>> corresponded to about 110vac input. (LV was 374v)
>> results:
>> HV ripple, scope ripple, Fluke
>> no cap 1014 8 v p-p 2.8 rms
>> 0.041 mf 904 8 2.5
>> 0.05 mf 875 6 2.2
>> 0.074 mf 856 4 1.2
>> 0.1 mf 866 3 0.7
>>
>> Make your own conclusions, it's interesting that the HV went up a bit at
>> 0.1 mf vs 0.074, with less ripple. I did not make actual capacitance
>> readings, just took stated values. The ripple's wavefor became more
>> distorted as the capacitance increased, at low values just a notch, but
>> pronounced double notches at 0.1mf.
>> I think I'd be happy with the factory value, or a bit more, in the 0.075
>> range.
>> thanks to all who responded,
>> 73,
>>
>> Al, W8UT
>> www.boatanchors.org
>> www.hammarlund.info
>>
>> "There is nothing -- absolutely nothing -- half so much
>> worth doing as simply messing about in boats"
>> Ratty, to Mole
>>
>> ______________________________________________________________
>> Collins mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/collins
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:Collins at mailman.qth.net
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>>
>>
>
> ______________________________________________________________
> Collins mailing list
> Home: http://mailman.qth.net/mailman/listinfo/collins
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Collins at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>
>