Sorry for delay, last day I have not e-mail notifications from forum.
For people it seems like cheating, but I never calculate anything in electronic circuits.
I think it is all on subconscious level. Simple I know it should be so, that's it.
Good article for power line filters design:
https://www.electronic.nu/2017/09/0...stallation-of-power-and-signal-lines-filters/

Thank you ... I guess I'm on my own, then. But thanks to your circuit, I can play with it, tweak it and adjust it ... see if I can get a basic understanding of what's happening.

 

I was not quite correct with filters. L1, C1 and L2, C5 are two absolutely identical filters.
Capacitor C2 works as power buffer with low inner resistance .
It provides instantly changing current for M1, M2 switch. Without C2 it is impossible through L1.
What is current, consumes by your 90VDC motor?
Here is article about filters for converting PWM signal to analog voltage:
https://www.allaboutcircuits.com/te...s-filter-a-pwm-signal-into-an-analog-voltage/
EDIT: And PWM frequency?

 

I was not quite correct with filters. L1, C1 and L2, C5 are two absolutely identical filters.
Capacitor C2 works as power buffer with low inner resistance .
It provides instantly changing current for M1, M2 switch. Without C2 it is impossible through L1.
What is current, consumes by your 90VDC motor?
Here is article about filters for converting PWM signal to analog voltage:
https://www.allaboutcircuits.com/te...s-filter-a-pwm-signal-into-an-analog-voltage/

I'm using two different 90VDC motors, one is 1/10 HP (1.1 Amps), and the other one is 1/6 HP (1.8 Amps)

 

Thank you ... I guess I'm on my own, then. But thanks to your circuit, I can play with it, tweak it and adjust it ... see if I can get a basic understanding of what's happening.

!
I felt "I'm on my own" in 5 years age, when i noticed that nobody can answer my questions.
From then onward I never ask people anything.

 

!
I felt "I'm on my own" in 5 years age, when i noticed that nobody can answer my questions.
From then onward I never ask people anything.

I was lucky enough to have a father who would answer all of the questions that he could answer. And those questions that he couldn't, he'd say to me: "I don't know, but why don't you find out by yourself?" ... and he said it with the purest sincerity... he was never annoyed by my continuous questionings. That's the moment when I fell in love with books.

 

For motor 1.1A or 1.8A values:
C1, C5 - 1uF
C2 - 10uF
L1, L2 -44mH SC-02-300JV COMMON MODE CHOKE 44MH 2A 2LN TH
https://www.digikey.com/product-detail/en/kemet/SC-02-300JV/399-10666-ND/4290732

 

For motor 1.1A or 1.8A values:
C1, C5 - 1uF
C2 - 10uF
L1, L2 -44mH SC-02-300JV COMMON MODE CHOKE 44MH 2A 2LN TH
https://www.digikey.com/product-detail/en/kemet/SC-02-300JV/399-10666-ND/4290732

And it doesn't matter if the motor is dc or ac split-capacitor?

 

And it doesn't matter if the motor is dc or ac split-capacitor?

Doesn't matter. Filters only convert PWM to AC sine or to rectified AC, without PWM pulses.
What is PWM frequency?

 

What is PWM frequency?

Anything I want. I'm generating it with an MCU, so it can be anything from a few hundred Hertz, up to perhaps 30 KHz. But I'm currently using 3.9 KHz, because it was easier to program.

So, the filter values that you've just posted work with ANY single-phase motor of up to 2 Amps?, and it doesn't matter if the motor consumes much less current? Say, will it also work for a motor consuming only 0.1 Amps?

 

Doesn't matter. Filters only convert PWM to AC sine or to rectified AC, without PWM pulses.

That answer is worth its weight in gold to me, by the way ... thanks!

 

I'm currently using 3.9 KHz, because it was easier to program.

Try to use frequency not lower than 10KHz, because it is boundary for this circuit.

 

So, the filter values that you've just posted work with ANY single-phase motor of up to 2 Amps?, and it doesn't matter if the motor consumes much less current? Say, will it also work for a motor consuming only 0.1 Amps?

Absolutely. It is ideal variac (if you will use circuit in post #134 ).

 

For motor 1.1A or 1.8A values:
C1, C5 - 1uF
C2 - 10uF
L1, L2 -44mH SC-02-300JV COMMON MODE CHOKE 44MH 2A 2LN TH
https://www.digikey.com/product-detail/en/kemet/SC-02-300JV/399-10666-ND/4290732

Danko, the choke you've recommended is of the two-line type. Here's a dumb question: Should I buy two chokes, and use one for each L1 and L2, or can the same choke be used for both lines? That is, line #1 of the choke to be used for L1 in the circuit, and the choke's line #2 for L2 in the circuit.

The reason I'm asking is because I don't know if there'd be any "coupling" between L1 and L2 with negative side effects if a single component were used for both inductors at the same time... instinct tells me that a single component is needed, but then again, instinct has failed me before.

 

The whole reason for the two windings on the choke is for both sides of the line. That way the saturation flux is cancelled and the thing does not produce harmonic distortion.

 

The whole reason for the two windings on the choke is for both sides of the line. That way the saturation flux is cancelled and the thing does not produce harmonic distortion.

That, I perfectly understand. But in Danko's design, the two windings are placed in different parts of the circuit.

 

If noise suppression chokes are not located at the inlet power points so that the current in the dual windings is not equal but opposite, then that benefit of core saturation prevention is not available, and so probably a single-winding choke would be the better choice, unless I have missed something. That happens occasionally. This string has a whole lot of comments and it is not hard to miss something.

 

Danko, the choke you've recommended is of the two-line type. Here's a dumb question: Should I buy two chokes, and use one for each L1 and L2, or can the same choke be used for both lines? That is, line #1 of the choke to be used for L1 in the circuit, and the choke's line #2 for L2 in the circuit.

Choke's coils should be connected in parallel. So you need one choke as L1 and second choke as L2.

The whole reason for the two windings on the choke is for both sides of the line. That way the saturation flux is cancelled and the thing does not produce harmonic distortion.

Yes, for line current winding are connected in series by counter way, therefore inductance for line current is zero, but for high frequency noise which run in the same direction in both wires of lines, winding connected in parallel, and have nominal inductance.

That, I perfectly understand. But in Danko's design, the two windings are placed in different parts of the circuit.

L1 and L2 are not coupled (no K parameter for them in simulation), so it is different chokes.

If noise suppression chokes are not located at the inlet power points so that the current in the dual windings is not equal but opposite, then that benefit of core saturation prevention is not available, and so probably a single-winding choke would be the better choice, unless I have missed something. That happens occasionally. This string has a whole lot of comments and it is not hard to miss something.

You are right, nobody knows, what is real maximal current below saturation in this chokes, with winding connected in parallel.
Seems I am trying to do my remote experiment, using TS money.

 

Seems I am trying to do my remote experiment, using TS money.

LOL! ... well, it will be money well spent, considering the potential results, and that this recommendation is coming from someone as experienced as you.

Anyway, so I need TWO separate components for L1 and L2. Here's another question then, can I use a double choke with half the recommended value, and connect its two coils in series so that their inductances would add up to the recommended value?

For example, in your previous post, you're recommending a 44 mH choke for L1 and L2. The way I'm understanding things now is that I'll have to buy to units, and then use only one coil of each unit for my circuit, leaving the other coil unconnected.

What if I were to buy a two-coil 22mH choke, and connect its coils in series so that their inductances would add up? Would that be acceptable, or would the fact that the two coils are sharing the same core be an issue? Also, if I were to connect them this way, would it be important which end of the coils are connected together, so that both are wound up in the same direction when connected, or should they be connected so that they're wound up in opposite directions? Or is this irrelevant?

Intuition tells me that I could connect the two coils in series, and that both should be connected with their winding in the same direction. For example, the end of one coil should be connected to the start of the next one. Like this:

​

 

Anyway, so I need TWO separate components for L1 and L2. Here's another question then, can I use a double choke with half the recommended value, and connect its two coils in series so that their inductances would add up to the recommended value?

Better solution for now:
Put behind common mode chokes.
Buy bunch of 3mH 3A inductors.
Use 15 inductors in series for L1 and 15 for L2.
Where they are:
30 inductors for $13.39
10 inductors for $4.91
10 inductors for $4.79
10 inductors for $4.91
All of them - free shipping.

 

Better solution for now:
Put behind common mode chokes.
Buy bunch of 3mH 3A inductors.
Use 15 inductors in series for L1 and 15 for L2.
Where they are:
30 inductors for $13.39
10 inductors for $4.91
10 inductors for $4.79
10 inductors for $4.91
All of them - free shipping.

Thank you, Danko! ... I'm currently considering winding my own (yeah, I'm that patient ) using the numbers output by this calculator. I'm guessing the amp rating of the coil thus produced will depend on its total impedance, and hence on the wire gauge used to wind it.