EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: sdouble on December 23, 2016, 04:56:03 pm
-
Hi Guys,
I guess , I'm in the newbie section. Perfect, here is a newbie question / ^-^
I power (really sensitive) preamps with a positive and a negative rail : + 6V and -2V.
I have a DP832 to provide me with the 2 voltages. I connect the DP832 to a 2 wire Lemo cable. THe cable is properly shieded. I fix the ground directly at the DP832 output : 2 poles of the PSU are connected to the braid of the cables. I need to mention that the cable and the preamp will operate in a crapy EM envronment : RF cavities, Primary and Turbo pumps everywhere.. really dirty.
THe cable will reach my equipment motherboard and I would like to proceed to a common mode filtering there.
I'm interested in the frequency range : 50kHz -> 5 MHz.
Now, my question :
should I use 2 common filters (1 rail and ground for each) and just one filter connecting direcly the 2 rails.
I guess that the 1st option is correct... but not really sure.
Any suggestion on the part I should use ?
THe current sucked is pretty low : less than 1 Amp.
-
One could use two separate filter (the more normal current compensated inductor) or one special common mode filter made for three lines. This a single core with 3 coils on it. They are available, but not very common and not so much versions to choose from.
-
In such a noisy environment you'll likely have noise on the cable shield as well, which means that you don't have a reference that you can eliminate incoming RF against. I would realize all communication with differential twisted pair signaling, use one additional twisted pair for the two supplies, add common mode chokes for all pairs at both ends, and add additional low pass filtering (C - Ferrite - C in a PI configuration) on all ends where possible.
-
thanks to both of you for your answers.
I'll look for a 3 line common mode filter. Could not find any yet, but we'll see.
The preamp outputs are filtered. RF should not be an issue.
However, I admit that moving to diff pairs could help.
Currently , my output are just single ended LMR240 coax. Those are good quality cables with braid and foil. In addition, I usually wrap the output cables with aluminium foils all together.
That way, I could never see on the scope any RF pick up.