[time-nuts] thunderbolt fault

[Bob Camp]

Hi

What I'm talking about here is a common mode choke on the +12 volt supply. They are pretty common items. Just about every TV or commercial switcher has one on the power line. Some switchers have them on the outputs as well. 

Two independent windings are put on the same core. The DC current / flux generated in one cancels the magnetizing flux from the other. That keeps the core from saturating. Thus you can get a lot of inductance in a small package. With a differential mode choke the thing that makes them *big* is core saturation. The obvious gotcha is that they will indeed saturate with very high levels of common mode noise. That's unlikely to be an issue here. 

The simple reason it would apply here is your earlier comment that the +12 is the supply to worry about. If the +12 and +5 are "garden variety" that would suggest a higher grade supply for the +12 source. As long as the currents are balanced, a common mode choke can be useful. 

No choke is going to help with DC regulation at the levels we are talking about here. What a choke is going to help with is audio and RF crud on the supply.  Things like an external three terminal regulator will help with differential noise on the supply. The same is true of bypass caps. What they will not help with is common mode noise. There are some exotic current source setups that will help with common mode, but a choke is normally a lot cheaper (and in most cases better approach).

At some point in the audio range the isolation provided by most regulators drops off. By the time you get to 100 KHz it's rare to find one that is doing a lot more than the bypass capacitors alone would do. If you know that the +12 is sensitive, doing a broad band isolation may indeed help things. Chokes of some sort are likely to be part of what you grab for the task. I've seen designs that go quite a way down the differential filter road when the real issue was common mode. 

There's no reason to throw away anything else here. It's just one more thing to consider adding to the design. Depending on exactly how things are or aren't set up it may or may not help.  

Bob


On Feb 26, 2010, at 10:44 PM, WarrenS wrote:

> 
> I don't know of any reasonable inductor thing that is useful to reduce High current ripple PS ripple.
> There where inductors, of the small bread box size, used in very early tube radios, But that was mostly just for low current B+ voltages.
> 
> Do you have a standard part in mind that will reduce 60 and 120Hz line voltage ripple by 20 dB or more at 1/2 amp rating,
> AND that does not hurt the voltage regulation with changing currents?
> What about one to reduce the DC voltage variations?
> 
> Your ground loop comment, sounds like you are talking about isolating the 10 MHz Osc ground to reduce ground loop problems and your comments are NOT about reducing the +12 volt PS ripple.  If so, Yea good idea, But  BIG difference.
> 
> As often happens on these post, there is misunderstandings somewhere and they tend to get way off the point,
> so have to go now and let others take the torch.
> 
> Original question was, Is the Tbolt's -12 volt supply sensitive or critical.
> SIMPLE answer, NO, just use a little common sense with it.
> 
> 
> Have fun
> ws
> **************
> Hi
> 
> It's pretty easy to get a common mode choke that will indeed break up 120 Hz ground loops. Often they are very low impedance. With most supplies the rectified line is what's coming through.
> 
> Bob
> 
> **************
> 
> On Feb 26, 2010, at 10:10 PM, WarrenS wrote:
>> 
>>>> Common mode choke filter does not get ride of LINE NOISE on the +12 V or LINE VOLTAGE sensitivities...,
>>> Please explain how a common mode choke on the +12 and ground does nothing to help keep the supply clean.
>> 
>> 
>> What size is your common mode choke filter?
>> To have any effect on 60 Hz PS ripple it would need to bigger than a (small) bread box .
>> And to help reduce DC type line voltage variations it would need to be bigger than a planet. (yea, Earth size)
>> Common mode filters are for HI freq, not 60 Hz OR DC.
>> But then, you already know that so I do not know why your comment???
>> 
>> ws
>> *************
>> Hi
>> 
>> Please explain how a common mode choke on the +12 and ground does nothing to help keep the supply clean.
>> 
>> Bob
>> *****************
>> 
>> On Feb 26, 2010, at 9:57 PM, WarrenS wrote:
>>> 
>>>> Lots of questions ....
>>> Same Simple answer.
>>> Make the +12 volts is as good as you can get it, For the rest any general purpose PS works fine.
>>> And a Common mode choke filter does not get ride of line noise on the +12 V or Line voltage sensitivities, both are important on the +12V
>>> 
>>> ws
>>> 
>>> ****************
>>> Hi
>>> 
>>> Depending on the supply setup, a common mode choke might also be a good idea.
>>> 
>>> The +12 runs the OCXO, so it's going to have an impact.
>>> 
>>> What about the +5 Volts? Obviously it needs to be crud free. Gross changes will impact the temperature of the unit. What about small changes? Is it running the maser reference for the DAC or does that come off the +12? Is +5 just a digital supply?
>>> 
>>> Lots of questions ....
>>> 
>>> Bob
>>> 
>>> On Feb 26, 2010, at 6:59 PM, WarrenS wrote:
>>> 
>>>> 
>>>> 
>>>> Yes, I Did that,   The +5 and -12 has NO effect on freq or operation of the unit except for really far out voltages.
>>>> I tested mine with the -12 from -2 to -15 and could see no effect in the e-11 range.
>>>> The -12 is used for the RS232 and the -Dac out so If you don't need the neg Dac out voltage or neg RS232  Drive, then it can pretty much be anything. But It should be kept in -7 to -13 range, and common sense means there should be little noise on it so it does not couple into other things. BUT the circuit it's self does not care what the supply is at or what is used.
>>>> 
>>>> The +12 on the other hand, Needs to be stable and quiet with no ripple.
>>>> I found using a +15 volt supply feeding a 12V three terminal will heat slinked regulator a good way to go to keep ALL the ripple out of the +12V
>>>> 
>>>> ws
>>>> 
>>>> *******************
>>>> 
>>>> Hi
>>>> 
>>>> Has anybody actually measured the supply sensitivity on the -12 volt line to see weather a 5 volt change makes any noticeable difference in the output frequency? The power on -12 is very low, so there should be negligible thermal impact from a change.
>>>> 
>>>> Bob
>> 
>> _______________________________________________ 
> 
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>