[time-nuts] need power trans. for HP3325

Dr Bruce Griffiths bruce.griffiths at xtra.co.nz
Sun Mar 25 22:28:57 EDT 2007


Didier Juges wrote:
> Dr Bruce Griffiths wrote:
>   
>> Didier
>>
>> Try using a Murata BNX002 filter between the switching supply output and 
>> the linear regulator input.
>> These should eliminate the High frequency noise (0.5MHz - 1GHz) from the 
>> switching regulator output leaving the linear regulator to deal with the 
>> low frequency noise.
>> However for these to be effective a low impedance ground plane is required.
>>
>> Worst case linear regulator design allowing for low mains input and some 
>> margin for the ripple from the rectifier certainly increases the series 
>> pass element dissipation considerably over that when the input to the 
>> regulator can be held within 0.1V or so with low ripple.
>>
>> The traditional preregulator employs SCRs in a phse controlled rectifier 
>> arrangement, the resulting low frequency output noise from such 
>> preregulators is easily dealt with by the linear regulator.
>>
>> Bruce
>>   
>>     
> Bruce,
>
> I work for a company that designs and builds switchmode power supplies, 
> so I have access to lots of power supply and filter components, and for 
> better or for worse, we are used to dealing with switcher noise.
>
> I will check the Murata filters though, I am not familiar with those.
>
> Most switching supplies operating up to 200kHz make significant noise up 
> to 30 to 50 MHz, and much less above that. Of course, if you design a 
> receiver, you may still be able to detect much lower level emissions up 
> to 300-400 MHz.
>
> Another pitfall of cheap switchers is line emissions, where noise is fed 
> back into the mains and may affect other equipment (by conduction or 
> radiation from the AC line itself). This is normally controlled by 
> regulations, such as FCC regulations here in the US for commercial 
> equipment, and via MIL-STD-461 for military equipment.
>
> It has been a long time since I have seen an SCR preregulator. They have 
> almost the same complexity as a switcher (from a parts count 
> standpoint), without any of the advantages, other than slightly lower 
> noise levels at higher frequencies. However, they are bad for line noise 
> (conducted emissions) and power quality (power factor, harmonic 
> distortion) and require oversized transformers, inductors and 
> capacitors, even compared to a normal linear supply. These are now 
> completely abandoned (except maybe at very high power levels, as used in 
> industrial processes and I believe in rail transportation) because they 
> can't meet the power quality requirements and they cost too much/are too 
> big and heavy.
>
> In the past, we have been using what we call linear-switchers (switcher 
> followed by a linear regulator) in some cases when the customer had high 
> expectations for cleanliness and efficiency, but nowadays, we can meet 
> all our customers needs with high frequency switchers, using good design 
> practices to minimize noise in the first place, and effective filters to 
> take care of the rest..
>
> Didier
>
> _______________________________________________
> time-nuts mailing list
> time-nuts at febo.com
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
>   
Didier

SCR's are still used in very high power (Megawatt) high voltage (0.5 
Megavolts or more) inverters where they replaced grid controlled mercury 
arc devices (some of these have been in operation for 40 years or more).

I am sceptical that any affordable switcher has an output noise as low 
as a well designed linear regulator (<10uV pp DC -20MHz).

Bruce




More information about the time-nuts mailing list