Multiple-Output DC-DC Converters with a Reduced Number of Active and Passive Components
Reviewer 1 Report
The organization of this paper must be improved. The reviewer cannot find what is the key point of this paper. It is more like an report rather than a research paper.
The length should also be reduced. It is too long.
There is no measurement result to support the analysis.
Thank you very much for your feedback and attention. We are aware that the paper outline is not a general paper style, because the paper proposes different multiple output converters. We want to verify each new converter with experimental results, so each of these converters has design criteria and experimental results: Section 2.1.2, Section 2.2.2, Section 3.1.2, Section 3.2.2, Section 4.3 and also the five-output converter prototype with closed-loop control: Section 6.3. These experimental results validate the analysis.
Thanks to your comment, to be more understandable to the readers, we changed the organization of the paper: The paper is divided into two parts. Part I proposes only new multiple output dc-dc converters and Part II explains how to model and control the most complex proposed multiple output converter. And also we decreased the length of the paper, now it is 8 pages less.
Reviewer 2 Report
I have read the paper titled “Multiple-Output DC-DC Converters with Reduced Number of Active and Passive Components” with great interest. This paper presents single-input, multiple-output DC-DC converter, with lower number of active and passive components. An alternative PWM-PFM-PD method is proposed. For five-output converter two of the outputs are adjusted using PWM in CCM, while the other two use PFM in DCM, and the fifth load is regulated by PD.
The paper is of interest, because it addresses a way to regulate several outputs from a single input. There are not many publications, where cross-regulation is treated and therefore it is a problem to solve and a good proposal. The paper also presents extensive experimental results and is well written.
This reviewer is struck by the large switching frequency variation with which the experimental results are obtained, for fly-buck two outputs converter, from 27 kHz to 420 kHz (Table 2), for forward-buck two outputs converter, from 45 kHz to 470 kHz (Table 3), for five outputs converter, from 85 kHz to 129 kHz (Table 8). Also, high frequencies the best efficiencies are obtained. The values of the passive components used are not clear in Tables 1 and 9. The authors state: “The efficiency at the lower switching frequencies is generally lower than the efficiency at the higher switching frequencies. This is because the ripple of iL2 decreases when fs increases. The conduction losses also decrease because of this reason”, but from my point of view, the use of high frequency presents a fundamental advantage: reduction of switching loss, on the other hand, the presents several problems: high dv/dt and di/dt of the power devices, EMI, high switching loss and stress. This limits the maximum switching frequency to a few tens of kilohertz (typically 20 to 50 kHz).
Finally, I suggest that authors consider reducing the length of the article, providing an indicative summary of sections 5 and 6.
Thank you very much for your positive feedback and attention. It motivates us to improve the paper.
There was a typo mistake in Table 1 and 9, because of different latex template, but it is solved now.
Nowadays, the switching frequency of low power dc-dc converters is increasing day by day. It is harder to design and implement these high-frequency dc-dc converters than those of the lower frequency as you said because of stress and EMI, but high switching frequency decreases the volume of the passive components, in other words, lower cost. MOSFETs and GaN switching components can reach up to 1Mhz and 10Mhz switching frequency, respectively.
Thanks to your comment, to be more understandable to the readers, we changed the organization of the paper: The paper is divided into two parts. Part I proposes only new multiple output dc-dc converters and Part II explains how to model and control the most complex proposed multiple output converter. Moreover, we summarized Section 5 and 6, we decreased the length of the paper, now it is 8 pages less.
Reviewer 3 Report
Multi-output converters becomes more popular so the topic is important. At this stage paper consists of many parts and it is not easy to read if there is not overall main idea. It is not clear which is main main message that authors want to give to reader. The suggestion is to make paper shorter, use possibly more colors in pictures, improve quality of the pictures, present main data from tables in a form of graphs. Usually duty cycle is indicated with D... Concentration on just flyback or direct forward converter and showing how additional outputs can be added and controlled seems better idea in my opinion otherwise it is difficult to for a reader with so many subsections of the paper.
Why in the figures isolated output is grounded? In Fig. 13 both connections to load are grounded - it seems as a mistake.
Thank you very much for your feedback and attention.
We agree with that D symbol is generally used for duty cycle, but also delta "δ" is the second option in power electronic papers. The reason why we used delta "δ" for duty cycle is that D symbol is used for the diodes. We need to mention diodes how to obtain additional outputs.
Thank you, It was a mistake in Fig. 13, but we corrected it.
Thanks to your comment, to be more understandable to the readers, we changed the organization of the paper: The paper is divided into two parts. Part I proposes only new multiple output dc-dc converters and Part II explains how to model and control the most complex proposed multiple output converter. Moreover, we summarized Section 5 and 6, we focused on the forward and flyback part more, we decreased the length of the paper, now it is 8 pages less.
Reviewer 4 Report
Complete paper on the subject of small converters five outputs, with a minimum of active and passive materials ranging from design to regulation. Few new theoretical elements but the interest is in the complexity of the system which is well treated.
Inaccuracies or failures to correct:
Figures 2, 5, 8 and 13: It is necessary to identify the voltages across the inductances: VL1, VL2, VL4, VL5 whose graphs are given.
Figures 2 and 5: With the current IL2 at the secondary of the transformer with a ratio n, the current of the primary is not IL2 but n.IL2
Figure 2: With the IL2 current at the secondary of the transformer, the primary current n.IL2 must be identified by a minus sign. The primary current should be (IL1 - n.IL2) (instead of IL1 + IL2) and identical to Figure 2, b (the drawing of the shape of the primary current in Figure 2, b is good). (also lines 100 and 102)
Figure 2, b: In the amplitudes of the voltages of VL2, why appears a parameter n2, it is normally n
line 97: (it can be increased by additional inverse windings  or by changing the overlap ratio of the secondary winding). I do not know the patent , could you explain in a few words how you increased the leakage inductance with a toroid transformer, because you have to use more wires and the position of the turns in opposition surely have importance.
In all the text, v2 is written in lowercase (example equ (1), line 107,…) or V2 (line 86, 99,…, equ (3),…) in upper case. We should harmonize the writing. (idem for v1, v3, v4 and v5).
Tables 1 and 9: problem for C and L unities: microFarad and milliHenry
Tables 2, 3, 4, 5 and 7: Specify the value of Vin.
Lines 107-108: The duty cycle 'delta' is kept the same for both these plots. There is a small change in V1 when fs changes because of cross-regulation. I do not understand how cross-regulation can intervene if the duty cycle is fixed? Also the same problem for me, line 144.
Figures 4 and 7: Specify the value of R2
Line 140: it is +5V/0,6A (not -5V)
Figures 10 and 11 : specify conditions: Vin, R3, no-regulation, ...
Tables 4 and 5: Specify the values of Vin and the frequencies.
Figures 18, 19 and 20: how are state variables initialized at the beginning of the simulation in matlab and in Spice? It would be preferable to initialize all state variables to zero, and after creating large transient regimes for Figures 18 and 19 and small variations for Figure 20, so as not to confuse the initialization of the simulation with the transient regimes.
From line 338, you use the letter k for a different meaning than at the beginning of the paper. You have to put two different names.
Comment Figure 24: The transient is not at 0.5ms but at 1ms
Thank you very much for your positive feedback and attention. It motivates us to improve the paper. We appreciate that you give the corrections.
We agree with that Figures 2, 5, 8 and 13 comments and we changed them.
Symbols were checked one more time, now they are consistent.
We specified more information about the experimental results in Tables 2, 3, 4, 5 and 7, Figures 4, 7, 10 and 11.
The other reviewers found the paper too long, this is the reason that we summarized Section 5 and 6, so we removed Figures 18, 19, 20 and 24, but we will publish the details in a Ph.D. dissertation more.
Line 97: First, a toroid is winded in one direction and then winded in the opposite direction. These windings will cancel each other respect to the magnetic field, only leakage inductance will be obtained. The patent has this idea.
Lines 107-108 about the cross-regulation: The cross-regulation is that a change in one output affects the other output(s), because of conduction losses, leakage inductance etc. For more information, you could check the paper from Dragan Maksimovic and Robert Erickson, MODELING OF CROSS-REGULATION IN MULTIPLE-OUTPUT FLYBACK CONVERTERS.
Reviewer 2 Report
From my point of view the paper has been improved enough. I believe that this paper is interesting for the JLPEA readers.
Reviewer 3 Report
The suggestions of the reviewers are takin into account.
Reviewer 4 Report
no new remarks