hi , i have a Lundahl dual coil choke 40H 60ma i ist possible to use the coil separately . like two choke of 20H 60ma for each channel ?
Hi!
This is not advisable since the two coils are coupled. You want independent chokes in a LCLC filter chain
Best regards
Thomas
This is not advisable since the two coils are coupled. You want independent chokes in a LCLC filter chain
Best regards
Thomas
hi thomas , i thought the lundahl use separate dual coils you can split or use in series or parallel .
Yes, you can use them in series or parallel. But not as two separate chokes in a LCLC since the two coils are coupled. This would defeat the purpose.
The two coils are meant to form one larger inductance choke when wired in series to one higher current choke when wired in parallel. But always to be used as single choke
The two coils are meant to form one larger inductance choke when wired in series to one higher current choke when wired in parallel. But always to be used as single choke
i beg you pardon for my ignorance , fortunately i have a second choke left from you 🙂
to be more precise do you mean that with my actual use i don't have a real 20H filter in my power supply or the left chanel can interact with the right chanel ?
to be more precise do you mean that with my actual use i don't have a real 20H filter in my power supply or the left chanel can interact with the right chanel ?
Last edited:
They can also be used as a common mode choke. See the data sheet.http://www.lundahl.se/wp-content/uploads/datasheets/1673.pdf
Hi!
When you use only one of the two coils you get a quarter of the inductance, so on your case 10Hy. If you use one coil on the left channel and the other on the right, you will probably still get decent filtering, but the two coils are coupled like primary and secondary on a transformer (albeit in this case probably a rather poor coupling)
Thomas
When you use only one of the two coils you get a quarter of the inductance, so on your case 10Hy. If you use one coil on the left channel and the other on the right, you will probably still get decent filtering, but the two coils are coupled like primary and secondary on a transformer (albeit in this case probably a rather poor coupling)
Thomas
If they are wound around the same core, they are not "loosely" coupled at all but **tightly** coupled.
Besides, as JonSnell said they *are* a transformer, and all ripple on the first half of the LCLC filter will be transferred to the second half, don't think you like that.
You really need 2 separate chokes.
Besides, as JonSnell said they *are* a transformer, and all ripple on the first half of the LCLC filter will be transferred to the second half, don't think you like that.
You really need 2 separate chokes.
hi , it"s not exactly that , i use one choke for the first L of my LCLC and the dual coil of the second dual coil choke split in two for left and right channel 🙂 it works fine but not at it should be referring to thomas opinion.
If you use half of this choke for l& r channel it will have the same effect as using both windings parallel and then separate filtering for l&r.
What I tried was 2 winding choke on 1 core for + and - power supply and that worked fine.
What I tried was 2 winding choke on 1 core for + and - power supply and that worked fine.
Hi, I should add that I have used the separate windings for right and left channel filtering, keeping care to make sure the phase is correct. It has worked well.
I only do it for filtering, as the last choke in an LCLC chain (or LCLCLC chain if you wish). It definitely IS important to wire it in correct phase.
I have had no issues doing this. My old o'scope shows nothing strange as compared to two separate chokes, and it saves space too... The sound is sublime...🙄
Yes, I agree they are "connected" since they share the same core material and are close to one another. However the windings are not intertwined.
Ian
I have not tried to do LCLC with a single Lundahl choke... This would not be so advisable, since Choke-Input duty is significantly different than Choke-Filter duty.
Perhaps this is where the confusion is.
Perhaps this is where the confusion is.
Hi!
If you use the two coils for left and right, you would probably be better off using the choke as intended in series connection and in common for both channels. This will give 4 times the inductance. Since the two coils are coupled the intention of having a separate independent left and right LC segment is definitely not met.
Thomas
If you use the two coils for left and right, you would probably be better off using the choke as intended in series connection and in common for both channels. This will give 4 times the inductance. Since the two coils are coupled the intention of having a separate independent left and right LC segment is definitely not met.
Thomas
I think we are talking about two different things...
Let me take the example of a Lundahl LL1638/8H
There are two 4H windings (connections 6+4, and 1+3). In series they are 8H. in parallel they are 2H.
But that is not what I am doing. I use them separately (yet in phase). Each winding is still 4H. In series, they are 8Hy which is more filtering than I need, but more importantly, results in greater serial resistance which I don't want.
I have had no problems with using Lundahl chokes like this, all my measurements are VERY good (as compared to separate chokes - even discrete channel to channel tests) and match my simulations VERY closely. No resitors in the power supply, which has super low serial resistance. The sound is of course sublime....
Anyway...
Last edited:
Sorry, Thomas has this exactly right and you need to take another look at how mutually coupled windings work. It is a transformer for anything that gets past the last capacitors in the filter network.. i.e, LF stuff, and you are throwing away 75% of available inductance that would otherwise be available. It's a compromise ultimately in low frequency isolation between channels and in potential rippled reduction - worst really of both worlds. One of those issues can be mitigated by doing what he suggests (both windings in series).