And I said there is no need for anything as the voltage across the choke is not reversing to negative ... maybe some people should refresh their knowledge about simple filters . This is not SMPS switching .
https://electronicsreference.com/analog/lc_filter/
https://electronicsreference.com/analog/lc_filter/
Last edited:
Thank you for all the very helpful information. I kind of like the idea of choke input especially when using SS diodes but it seems that PS input swinging chokes are a more scarce item than regular PS filter chokes.
There is no off-cycle in a full wave rectifier. Current comes from one phase or the other. But some circuits still use a ground diode to cover the dead time.
It is the same for half wave , not the missing cycle is important but the waveform
As a principle people should understand that without a physical switch to interrupt a DC current it is not possible for an inductor to switch polarity
As a principle people should understand that without a physical switch to interrupt a DC current it is not possible for an inductor to switch polarity
For a choke input filter, having a valve diode between the bridge and the choke does not change the hard commutation of the bridge diodes - the valve diode is effectively just adding to the series resistance of the choke.
I don't recall seeing any choke input filter schematics in amps showing a 'ground diode' - can you link to a schematic or amp ?
Hammond had several swinging chokes in their product line at one time.
They would have been used in the PS for Class B amps.
I like the so called ground diode & have included it in a couple of bench PSs where C or L input can be selected.
Especially if a B+ standby switch is involved.
Chokes for such duty need to have flux density headroom for the accounted AC voltage swing across. They will also radiate more EMF.
Due to the AC voltage they will as well vibrate more.
Swinging choke is one that saturates easily from Idc. Here is what happens.
1. Idc increases due to load requirements.
2. Choke flux density increases, reaches BH knee, chokes saturates.
3. Permeability decreases
4. Inductance decreases.
5. But because L vs I decreases linearly and critical current is linear, the power supply remains in choke input. Swinging chokes are advantageous for, for example, class AB amplifiers. I'm using an in-house swinging choke input for my LM1875 amplifier.
Due to the AC voltage they will as well vibrate more.
Swinging choke is one that saturates easily from Idc. Here is what happens.
1. Idc increases due to load requirements.
2. Choke flux density increases, reaches BH knee, chokes saturates.
3. Permeability decreases
4. Inductance decreases.
5. But because L vs I decreases linearly and critical current is linear, the power supply remains in choke input. Swinging chokes are advantageous for, for example, class AB amplifiers. I'm using an in-house swinging choke input for my LM1875 amplifier.
Hi,
I'd like to use a smoothing choke as an input choke in this over-the-top PS, Triad model C-14X 200ma 6H. As you can see the ripple is extremely low and I have a boatload of chokes and a hankering to use them. The current across L1 is RMS 50ma 30ma mean and maxes never above 86ma as the model runs. I'm assuming this circuit meets the minimum load and inductance for a choke input to be working properly? I chose a 200ma choke to make sure I had enough headroom for both the DC and AC load as I only need 50ma across it. Is that enough room to make using a smoothing choke just fine here? I've been on a few ham radio boards and many say smoothing chokes can work just fine for choke input. But I was also just reading Merlins new book where he mentions problems using a smoothing choke. What are we to do when the market no longer makes dedicated input chokes? If derated enough shouldn't it be possible to use one? How am I doing here, this C-14X will be potted BTW.
I'd like to use a smoothing choke as an input choke in this over-the-top PS, Triad model C-14X 200ma 6H. As you can see the ripple is extremely low and I have a boatload of chokes and a hankering to use them. The current across L1 is RMS 50ma 30ma mean and maxes never above 86ma as the model runs. I'm assuming this circuit meets the minimum load and inductance for a choke input to be working properly? I chose a 200ma choke to make sure I had enough headroom for both the DC and AC load as I only need 50ma across it. Is that enough room to make using a smoothing choke just fine here? I've been on a few ham radio boards and many say smoothing chokes can work just fine for choke input. But I was also just reading Merlins new book where he mentions problems using a smoothing choke. What are we to do when the market no longer makes dedicated input chokes? If derated enough shouldn't it be possible to use one? How am I doing here, this C-14X will be potted BTW.
IMHO Today choke input is special because it's big and heavy and takes a lot of chassis real estate.
If you use a solid state rectification scheme, you can use larger caps which usually supplant the need for a big coil...
My first amp used a Hammond 193Q which is only a DCR of 53R and gives far better regulation than using cap input with the 47uF cap specified as max for a 5U4GB... However, I now use HER308 and much larger caps... the R where the choke was is only 5R now because I can use many times more capacitance 🙂
The amp I'm listening to right now has HER308 -> 565µF -> 5R -> 660µF - 5R -> 660µF for a 50W stereo triode amp... My first design was a typical "book" design of 5U4 -> 47µF -> 10H -> 47µF...
If you use a solid state rectification scheme, you can use larger caps which usually supplant the need for a big coil...
My first amp used a Hammond 193Q which is only a DCR of 53R and gives far better regulation than using cap input with the 47uF cap specified as max for a 5U4GB... However, I now use HER308 and much larger caps... the R where the choke was is only 5R now because I can use many times more capacitance 🙂
The amp I'm listening to right now has HER308 -> 565µF -> 5R -> 660µF - 5R -> 660µF for a 50W stereo triode amp... My first design was a typical "book" design of 5U4 -> 47µF -> 10H -> 47µF...
Last edited:
A swinging choke utilises a non-uniform air-gap, whereby portions of the distributed air gap reach higher flux levels (due to smaller gap length) than other portions (which have a larger gap) as the choke current cycles in level. In its simplest form it's like the core has two steps of air gap, one longer than the other. The effective inductance transitions from a higher level to a lower level as choke current cycles up in to the core saturation region.
PSUD2 doesn't simulate a swinging choke, and imho any sim of a choke-input filter supply needs to be taken with a lot of caution, as the choke model being used is just a simple ideal inductor. The user may not have measured the choke (or relied on manufacturer data) to obtain a good awareness of what inductance is being exhibited by the choke in the application.
The region of the operating cycle that is of concern is when the choke current approaches zero (ie. not when the core has max current and is entering its saturation region), as that is when the diode is again forced to commutate off and on (in addition to the normal commutation when the diodes alternate conduction during each half cycle of mains).
Maybe I'm not finding a good enough datasheet for the triad C-14X, but it has no definition of voltage conditions for the inductance spec, only a DC current spec.
PSUD2 doesn't simulate a swinging choke, and imho any sim of a choke-input filter supply needs to be taken with a lot of caution, as the choke model being used is just a simple ideal inductor. The user may not have measured the choke (or relied on manufacturer data) to obtain a good awareness of what inductance is being exhibited by the choke in the application.
The region of the operating cycle that is of concern is when the choke current approaches zero (ie. not when the core has max current and is entering its saturation region), as that is when the diode is again forced to commutate off and on (in addition to the normal commutation when the diodes alternate conduction during each half cycle of mains).
Maybe I'm not finding a good enough datasheet for the triad C-14X, but it has no definition of voltage conditions for the inductance spec, only a DC current spec.
Its only a .047 uf capacitor so it was my understanding that the circuit shown is operating still as primarily choke input. Some things I've read suggest a small capacitor there for oscillation.
Y
Think I ran some amplitude vs time sims on something similar a while back.
It looked very ragged when current was below the critical. 😀
You are correct, I goofed. The result of a quick look!
Think I ran some amplitude vs time sims on something similar a while back.
It looked very ragged when current was below the critical. 😀
You might want to include a small resistor in series with that cap. It won't show in PSUD, but will help to damp the transformer's ringing following diode transitions. Stuff gets everywhere. Maybe something in the 100 Ohm to 1K Ohm region.
All good fortune,
Chris
All good fortune,
Chris
I see, psud wont do that but when I get it on the breadboard I'll note this.
One of the advantages of a choke input powersupply is the transformation of the sound of your preamp or power amp. Compared to a clc power supply the LC sounds way better.
Trying to understand this as a Newby. So a swinging choke swings through a range of henries based on the current fluctuating. The fact that the current is totally unsmoothed pulsed DC is the current swinging on the peaks that varies the inductance? Or is it swung because the load side varies? But my application has a stable load class A pre Amp. I see swinging chokes are spec'ed using two Henry numbers with milliamp pairs. One pair has the low Henry with the high MA and the other pair has high henries at low MA. In my application the load side is constant. What else would cause it to swing other than the pulsed DC? Trying to understand why a non swinging choke won't work as well even if the load is constant, nothing else is moving here except the pulsating DC.
Plus I really want to try one for fun! Now in your experience can I use a non-swinging choke? Since they really don't make swinging chokes off the shelf anymore. Assuming I have enough load so things are not running away. I'm happy to breadboard it to death until its working, but don't want to embark that journey if I needed a swinging choke all along and did all that work for naught.