Hello all,
I've been reading up on snubbers for power transformer secondaries, including the work of John Swenson and Jim Hagerman. One bit isn't clear to me though - when using a center tapped HT transformer, should the snubber RC network be run from one end of the secondary to the other, or should there be two RC networks, each running from one end of the secondary to the center tap? After plowing through several web and DIYAudio searches I have found examples of both; is there a standard "best practice"?
I've been reading up on snubbers for power transformer secondaries, including the work of John Swenson and Jim Hagerman. One bit isn't clear to me though - when using a center tapped HT transformer, should the snubber RC network be run from one end of the secondary to the other, or should there be two RC networks, each running from one end of the secondary to the center tap? After plowing through several web and DIYAudio searches I have found examples of both; is there a standard "best practice"?
Serge, thanks for the reply.
I usually do put an RC network on the primary, to remove noise coming from the mains.
What I'm talking about in this post, though, is a snubber on the secondary meant to absorb any ringing that occurs between the transformer secondary and the rectifier. There has been quite a bit posted on this subject but I can't seem to find a clear answer regarding how best to do it for CT secondaries; it's quite clear with a bridge rectifier that it can only be placed completely across the secondary winding. Most use a 330ohm resistor with a .022uF cap for this.
I usually do put an RC network on the primary, to remove noise coming from the mains.
What I'm talking about in this post, though, is a snubber on the secondary meant to absorb any ringing that occurs between the transformer secondary and the rectifier. There has been quite a bit posted on this subject but I can't seem to find a clear answer regarding how best to do it for CT secondaries; it's quite clear with a bridge rectifier that it can only be placed completely across the secondary winding. Most use a 330ohm resistor with a .022uF cap for this.
A single network across the entire secondary will suffice. Even if your rectifier is FWCT as opposed to a 4-device bridge. As long as you have the correct values for R and C, (and a scope, of course) it will work.
Personally, all articles attempting to predict RC values have not helped me. The best way is by experiment with a properly loaded circuit. You start by identifying the smallest C that makes a change, and pick the largest R that effectively snubs. Trial and error will quickly show you which value needs to go up or down from there. Measuring ringing frequencies can help to pick ballpark values, but at the end of the day there is an ideal set for optimal performance. Too much C (or too low R) just consumes excess power, and too small C (or too high R) does not snub the transients.
Personally, all articles attempting to predict RC values have not helped me. The best way is by experiment with a properly loaded circuit. You start by identifying the smallest C that makes a change, and pick the largest R that effectively snubs. Trial and error will quickly show you which value needs to go up or down from there. Measuring ringing frequencies can help to pick ballpark values, but at the end of the day there is an ideal set for optimal performance. Too much C (or too low R) just consumes excess power, and too small C (or too high R) does not snub the transients.
"Snubbers" are used to remove ringing in switching devices under inductive loads, to make a lossy tank and discharge leakage inductances in transformers or motors, they don´t remove any line noise, you must use a line filter to do this.
The snubber is there to re-route leakage type current flow from the diode switching. The best route is to place the snubber as close to the section of winding that is undergoing the switching - in the normal CT secondary case, this is from each end of the secondary to the CT, and as close as possible to the transformer, rather than out at the diodes (but in practise you don't typically get any option). The loop area of the secondary to CT helps radiate any noise, so best to twist any such secondary leads, and not pass them near to sensisitve inputs (as you would do for heater distribution).
The two secondaries (or halves of a CT secondary) are magnetically coupled together, so a snubber on one alone will reduce ringing on the other.
If you want "agreement" on technical issues then you either need to find a forum where everyone thinks alike (but this does not guarantee truth) or be able to filter out those who don't understand the problem and then take a consensus from those remaining.
If you want "agreement" on technical issues then you either need to find a forum where everyone thinks alike (but this does not guarantee truth) or be able to filter out those who don't understand the problem and then take a consensus from those remaining.
I didn't say I "wanted" agreement, I said there doesn't appear to be any in this case. Of course it would be nice if there was one sure answer, but like many other topics in life that doesn't seem to be the case here.
My interpretation of the consensus in this case is that one snubber across the entire secondary is adequate, and one across each half to CT may be a little better; perhaps a case of diminishing returns. Being somewhat obsessive, and given the low cost of the parts involved, I think I'll try the latter.
Thanks to all!
In a regenerative receiver I made, I put a snubber across entirely secondary winding of 1K and .15uF on a 170 + 170V to the plates of a 6X4. Placing across all secondary you are snubbering all leakage inductance. As DF96 tells, there are magnetically coupled, but not only inductive, also capacitivelly, and the linkage is not 100%, so there exists a leakgae between secondaries itself.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.