+380.000 other summaries
A unique study tool
A rehearsal system for this summary
Studycoaching with videos
Remember faster, study better. Scientifically proven.
Summary - 28V Space bus Solar Array Regulator Based on Converters with Transformer and Self-Driven Synchounous Rectification
1 Introduction Text
The most common topologies for main bus regulator used in European satellites are non isolated topologies. Buck boost are the most used ones. The main reasons for this: robustness and efficiency. Robustness because satellites can not be repared while in orbit, and efficiency to reduce the size of solar array's and to simplify the thermal system.
Another key point is failure and protections. The design of the power system has to be Sinlge Point Failure Free (SPFF), which means that any single failure on the system cannot lead to any loss of performance. Therefore, if you want a 200W power system, three 100W converters are placed in parallel. If one should fail, then you are still able to deliver 200W.
Furthermore, in conventional buck and boost converters, if the main switch fails in short circuit it will connect the solar array to the main bus permenantly. To cope with this issue, another switch is needed in series with the main. This also applies for the diode. Obviously, this decreases the efficiency and the complexity is higher.
An alternative based on topologies with a transformer could be preferable, since it is isolated is doesn't need extra switches. The use of Synchounous Recification and modern MOSFETS's can though boost the efficiency to levels that might be very close to the traditional ones.
Why could the uses of topologies using transformers be preferable in satellites?because a topologie that uses a transformers doe not need extra switches to protect is from shorting, because the transformer in isolated and only works with an AC voltage.
2 The Half Bridge Topology For Space Power Systems
As for protection in a Half-bridge topologie: The input stage is self protected since it interfaces the solar array with two switches in series.
Regarding the output stage, two options are possible: A four diode rectifier can be used and no extra switch is needed, the output rectifier is based on a centrer-taped winding in which an extra switch is needed.
The preferred choice is the four diode rectifier since the converter can then be used without extra switches.
The half-bridge topology is also favourable because the fact that it can be easily adapted to any solar array design by changing the transformer turns ratio.
The remaining problem is now is efficiency.
The transformer has two effects:
1. it adds some losses due to its own operation.
2. it degrades the operation of the switches since it adds some leakage inductance on the switching loops.
This will cut down the efficiency to 1 or 2 point easily.
The leakage inductance can be minimized by implementing planar transformers, but there is no way to remove the losses of the transformer itself. However, the efficiency can still be boosted a bit by using synchrounous rectification.
What is the disadvantage of a Half-Bridge over a Buck or Boost converter?Its efficiency is lower due to the transformer its leakage inductance and its own operation.
Why is the Half-bridge converter a favourable converter in space-applications?Because it minimizes the need for protection such as extra switches and diodes. It already has two switches in series and the secondary side is isolated by a transformer. Furthermore, it can easily be adapted to any solar array just by changing the turns ratio
How can the efficiency of the Half-Bridge be boosted?First, planar transformers can be used to minimize the leagake inductance. Secondly, by using Synchrounous Rectification.
3 Synchounous Rectification in Bridges
The voltage drop across a MOSFET can be lower than the voltage drop across a diode. However, for very high current levels the losses on a MOSFET will be higher than in a diode. Obviously, the switching losses also have to be taken into account in the global computation.
In the case of the Half-Bridge, during the dead times on the transformer, the voltage is zero and hence, SR is not very effecient because the body diodes conduct the current ( which are very poor diodes ). The following chapters will present a method to overcome this problem.
4 Proposed SDSR system
To discharge, two windings coupled to the main transformer are connected between the gates and an off-setting voltage source made with a simple capacitor and a zener.
When the SR MOSFET has to be on, the Voff voltage is higher than the gate voltage Vgs and the diode Doff will block it.
When the SR MOSFET has to be turned off, the voltage Voff will be negative (or zero) because the winding voltage changes its polarity. Then, he diode Doff will conduct, discharging the gate to the capacitor voltage minus the winding voltage.
The objective of this method is to have some voltage at the gate, even during the dead times.
By looking at the waveforms of the half-bridge, one can use a suited waveform to turn the SR off, being the transformer waveform when using the right polarity.
The key issue is then to obtain a driver that lets the SR inherently on for the rest of the period. This can be achieved by using the output inductor waveform.
With this method, a winding can be coupled to the output inductor to adapt the voltage level to a value acceptable for the SR gate as seen in Fig, 5a. As the MOSFET gate is capacitive and in charged through a diode, once it is charged it will keep that voltage indefinitly unless another circuit discharges it.
The coupled winding at the output inductor, drives turns both switches ON. Why is this not a problem?Actualy, it doesn't turn both switches ON since one is already ON. This is not a problem because that is exactly what you want, to get rid of the dead times. The main switches will make sure that the current will flow through M1 and M2 will not be simultainously.
Why can't you also make use of the output inductor when switching OFF?Because in that case, you don't want them to turn off at the same time.