Good Day,

I like to discuss this circuit. It provides pulsating current to the electromagnet. I used a buck converter, a mosfet switching module and a capacitor bank.

This is the Mosfet switching module.

https://zbotic.in/product/5-36v-swi...L7-DXSK7NUbNSbSbloIzSPcGeC61mCt4aAtzmEALw_wcB

This circuit provides 3.75 Amperes at 7.48 Volts from the buck converter. What I wanted to do is to control the vertical down limit of the current which is zero now while switching OFF time period. I checked the Mosfet module input and output pins. The resistance is zero between the "+" input and "+" output. Whereas the "-" input and "-" output terminals are open when the PWM signal reaches 0. So I connected a power resistor of suitable value 1.6 Ohms connecting the "-" input and output terminals as shown in the below circuit.

Now I can send 2 Ampere current through the load when the PWM signal is zero. So the pulsing waveform is passed through the load as calculated without hitting zero.

Now I like to try something more. I like to connect another buck converter 2 in parallel to the buck converter 1 as shown below. So I can manually adjust the vertical max and min value of the pulsing current.

My doubt is would this work out? I want to discuss it here before trying it out and take advice. Two buck converters are in parallel one would supply the current always to the load. The other one will cut in and cut out controlled by PWM generator. So I expect the peak value would be the sum of two currents during the ON time period and during the OFF time period only buck converter 2 supplies the current to the load. I have placed diodes at the buck converter 2 outputs.

These are the buck converters I am using one is a high power one and other one is a low power one. In case of parallel connection I should use the low power mosfet for switching.

https://www.xcluma.com/20a-300w-dc-...qIDwfa9c4iY10S_HjZeqvYh4mdoH28rxoCwpsQAvD_BwE

https://robocraze.com/products/5a-d...sfojt8U8MAEwqTAyheLFW1eLyvgkNEJhoCE3MQAvD_BwE

 

Buck converters in parallel? IMHO -- Bad idea. How would the respective feedback and compensation networks be able to sort out the conflict between two power stages going in opposite directions?

 

Buck converters in parallel? IMHO -- Bad idea. How would the respective feedback and compensation networks be able to sort out the conflict between two power stages going in opposite directions?

Yes possibly that is a bad idea that is why I didn't try it out and first ask about it since my knowledge about buck converters is less. I was considering it as two voltage sources in parallel but as you mention the feedback wouldnt possibly work correctly.

So I will go about with the resistor option and may be try out a variable power resistor.

Thanks

 

To vary the load power by switching it is more common to add sources IN SERIES, changing the voltage applied to a resistance. What it looks like to me is excessively complicated. There is a scheme to have multiple ISOLATED voltage sources in seriesusing only a single pole switch to add them.
so now questions. 1.How important is maximum efficiency? and #2 How good is the TS understanding of circuit theory??

 

so now questions. 1.How important is maximum efficiency? and #2 How good is the TS understanding of circuit theory??

Maximum efficiency is not important now I am only doing the testing now by varying the current. I did it by swapping the different values of resistors manually in parallel and removing as needed (since a variable power resistor is way too expensive). Now I have settled at an optimum value of current that needs to be present without letting the waveform hit zero. So I am not in a requirement to vary the voltages and currents anymore for now. Thanks for your response.

I have basic understanding of Circuit Theory but I didnt pursue that field I worked as Electrical Service Engineer sailing on ships for 11 years before quitting so I lack knowledge on design concepts.

 

With very few exceptions, clients start out with a description of what they want to achieve, rather than a description of how to achieve some unstated goal. They do sometimes explain how some method has problems, issues, or costs too much, or is not fast enough. But always presented at the beginning is the statement of what the purpose or goal is.
What sort of comes across is that changing the strength of an electromagnet is the goal, while what we are asked about is how to put current sources in parallel.
For starters, it is not done that way because it does not work that way.

 

Exactly, MB.
What you describe is a case known as the XY problem.

https://xyproblem.info/

 

I see it here all the time: The favorite is when there is a circuit problem and somebody posts just that very small part of the circuit with a dozen lines leaving and they are asking why one thing does not work right.
I have not yet replied and told them that if they know where the problem is, why are they asking ME???
That would be a bit rude, and I try to avoid that at all times. So I ask for the information that is needed, and usually need to remind folks that without knowing the whole situation, the best they will get is blind guesses.
In this magnet question, the way to increase the current is to increase the voltage. BUT while that is required, it is not sufficient.
Simply, to have more current you must have more voltage to push it thru. But then, also, to have more current the ability to deliver it must also be present.
At that point things DO get complicated. It reduces to the math that to have the voltage that pushes the current you do need MORE POWER capability.

 

What sort of comes across is that changing the strength of an electromagnet is the goal, while what we are asked about is how to put current sources in parallel.
For starters, it is not done that way because it does not work that way.

MisterBill2 and schmitt trigger,

Apologize for any misunderstanding caused by not being clear in my posts. However I only wanted to know if buck converters can be operated in parallel that was my original question in this thread for that inductive load.

The reason is I want to control the vertical part of pulsing waveform of the current through the electromagnet which are the up and down limits using the pots on the buck converters. So I wanted to try out two buck converters in parallel but since that wasnt a plausible solution due to feedback problems I gave up on that idea and went with other option of varying the resistance values to change the current waveform. I do understand your logic regarding the basics of Electrical Engineering about Voltage and Current and how they work and their application here.

I want to learn from others here before trying out anything which could result in losses. Again, I do not intend to cause any confusion.

 

Parallel operation of any feedback regulated supply is always more complex, as you mention. And it will not cause more current to flow unless the voltage is increased. The current a load will draw depends on the load resistance, not on the supply capability. Of course, if the load resistance is low enough to demand more current than the supply can deliver, then the actual current will be limited by the supply capability.
The references to "vertical" tend to be a bit confusing.
What I am guessing is that the goal is to be able to have the current in the electromagnet change to different values, including zero, at different times, according to some external control. That is entirely possible, and there is already a (current) thread that has discussed that and found a solution.
The solution is not trivial but appears simpler than what has been presented in this thread.

 

The references to "vertical" tend to be a bit confusing.

I should have posted this waveform drawing before mentioning about vertical. That was something I missed.

Yes intially I was using higher voltage sources and higher resistances to pass higher load current. BobTPH here adviced to use the buck converter to prevent the wastage of power due to resistors since that would provide higher current at a lower voltage. So I used a buck converter and saved a lot of power being wasted. When I wanted to pass pulsing current through the load I used this MOSFET switching module to do it but the downside vertical limit of the current waveform hits zero and I want it to be non zero. I connected a resistor across the MOSFET switch as in post #1 so the current does not go to zero in the waveform. But I wanted to increase and decrease this current value so I wanted to know if two buck converters can be connected in parallel. Since that isn't a good idea I tried changing the current by changing the switch bypass resistor's values manually.

 

I presented a higher efficiency scheme in post #4.
Switching the connection to different points in a series stack of voltage/current supplies can allow a large range of current settings without any power being wasted in resistors.

 

I presented a higher efficiency scheme in post #4.
Switching the connection to different points in a series stack of voltage/current supplies can allow a large range of current settings without any power being wasted in resistors.

Yes I read this idea it is good and it increases efficiency. But here in my case I am using a switching device and the max limit of the waveform I can control using the main buck converter pots. I need to control the lower limit of the current when the Mosfet is not conducting. That is why I didnt know if this would work out here. Also by changing the resistors I arrived at the required lower limit current value and I dont need to change it anymore. But I am curious about how you switch those Voltage sources in series to increase efficiency. In my setup now since I have arrived at the max and min values of current I wouldn't change it anymore most probably but I could be doing it in future. But I will need to change the frequency for that I decided to use MCU.

Thanks.

 

OK, I see the question clearly now.
My switching scheme always used mechanical contacts in an actual physical electro-mechanical relay device. That avoids the requirement of isolating the control circuit. BUT it does not remove the requirement for the individual voltage sources to be isolated from each other except to allow "stacking" them in series. The down side is that it does tend to slow the maximum rate of change of the output.
If you visualize it as a series string of batteries and the tap points are the connections between the individual batteries, that will provide a correct picture of the concept, and where the isolation would be required.

 

If you use the buck regulators IN CONSTANT CURRENT MODE it should be OK to connects the outputs in parallel.
BUT as the negative of the buck regulators is common you would need to use a switcing module that switched the positive line.
One other point is you show the required output waveform with fast rise and fall transitions. The capacitor board would slow down the transitions. I do not understand the function of the capacitor board.

Les.

 

If you use the buck regulators IN CONSTANT CURRENT MODE it should be OK to connects the outputs in parallel.
BUT as the negative of the buck regulators is common you would need to use a switcing module that switched the positive line.
One other point is you show the required output waveform with fast rise and fall transitions. The capacitor board would slow down the transitions. I do not understand the function of the capacitor board.

Les.

The capacitor board was added to reduce oscillations to the load. Since both the buck converter and the mosfet module are switching at different frequencies.

I want one buck converter supplying current continuously to the load and the other to cut in and cut out using pwm and mosfet module. Buck converter 1 will be cutting in and cutting out while buck converter 2 supplies power to the load continuously. When the currents add up they reach the top of the waveform and when buck converter 1 is removed by the pwm signal to the mosfet buck converter 2 decides the bottom of the waveform.

 

When you talk about PWM do you mean chaging the duty cycle of the waveform shown in post #11 ?
You have no units of time shown on the waveform in post #11 so we have no idea of time. (We need to have some idea of this as it might be nS or years.) The buck regulator may not like it's load changing rapidly.

Les.

 

When you talk about PWM do you mean chaging the duty cycle of the waveform shown in post #11 ?
You have no units of time shown on the waveform in post #11 so we have no idea of time. (We need to have some idea of this as it might be nS or years.)

Les.

Actually that waveform isn't accurate I just gave a rough example. I use pwm of different frequencies all with 50 percent duty cycle that is constant only I change the frequencies every one hour to few minutes. I use 0.1 Hz, 7 Hz, 14 Hz, 21 Hz, 28 Hz, 35 Hz and 42 Hz pwm signals to the mosfet switching module. When I am using 0.1 Hz it will be ON for 5 Seconds and OFF for 5 seconds and repeat. The current will be in its max value and min value for 5 seconds.

 

This is so very similar to another thread that this TS should study how the other has solved the same problem.
OR is this the same TS using a different name???

 

This is so very similar to another thread that this TS should study how the other has solved the same problem.
OR is this the same TS using a different name???

No the other thread was about changing the frequency of the current waveform. This thread is about changing the magnitude of current. Both are different.