Advertisement
Articles
Advertisement

Choosing the right IGBT Trade-Offs to maximize Motor Drive performance

Tue, 09/08/2009 - 5:29am
Vijay Bolloju, Jun Yang , & Elk Kabaker, IGBT Applications Engineering group, International Rectifier

 

EKakaber_IMG_4013-webRecent advances in IGBT technology have given motor drive designers more options to consider for system development.  Different specification conventions of various vendors complicate the design process further.  Conventionally, system efficiency has been equated directly with VCEON.  This metric is important for calculating the conduction losses of an IGBT but leaves out switching losses completely.  This article aims to show how IGBT switching losses effect the performance of motor drive inverters and should be considered in the design phase.

JYang-webIGBT losses in an inverter are a combination of conduction and switching losses. The VCEON contributes to the conduction losses and ETS contributes to the switching losses of the device.  When choosing an IGBT for an application, both VCEON and switching energy (ETS) specifications must be taken in to consideration. IGBTs are minority carrier devices and as a result, they have a trade-off between VCEON & ETS [1]. A fast IGBT tends to have lower ETS at the expense of higher VCEON and vice versa. In order to maximize the efficiency of the inverter, the IGBT trade-offs need to be matched to the PWM switching frequency range of the inverter.   For this analysis, two 1200V trench IGBTs are evaluated, one optimized for lower VCEON and the other vbolloju-weboptimized for lower ETS.  Both devices offer 10 us minimum short circuit capability.  The parametrics of these devices are shown in the following table.

Two key operating characteristics should be evaluated in designing a motor drive inverter system.  The first is maximum current or power delivering capability.  Most inverter systems are designed to deliver an overload current for a set amount of time.  The second key consideration is efficiency.  Beside the benefit of wasting less power, higher efficiency can also reduce system cost and size.  Heatsink size directly correlates to power dissipation and is therefore dependent on the efficiency.  There are also regulations and incentives for meeting certain efficiency standards. 

table1-ir-0910One effective way to evaluate the maximum current carrying capability of different IGBTs in an application is to compare RMS current capability vs. switching frequency.  This is exemplified in figure 1 for the two devices being evaluating in this article.  This graph is generated by determining the maximum power dissipation of the IGBT based on DBC Module with Sine PWM control. The Case temperature is set at 100 ºC & Junction temperature at 125 ºC.  Some IGBT manufacturers provide similar graphs on their datasheets to assist designers. The results shown in Figure 1 are as expected; below a certain frequency, the slower fig 01-ir-0910device offers higher current carrying capability and then beyond the crossover point the fast device performs better.  This shows the inherent limitations of each IGBT.  A system requiring 70A RMS current capability operating at a frequency of 2 kHz benefits if built using the slower IGBT.  Conversely, a system requiring 50A RMS current capability operating at a frequency of 8 kHz benefits if built using the faster IGBT.  

Determining the efficiency of the inverter amounts to calculating the losses for a given amount of current. Because of the switching component of the losses, the efficiency will depend on the frequency of operation.  The graph in Figure 2 shows the IGBT losses vs. RMS output current of these IGBTs in an inverter system at 4 kHz & 8 kHz.  The overall system efficiency will be application dependent and relate to the typical load cycle of the inverter.


The graph in Figure 2 also shows that at 8 kHz operation, where switching losses have a more significant contribution, IGBT1 (the faster device) has considerably lower losses (Higher efficiency) than IGBT2.   The less intuitive point fig 02-ir-0910demonstrated in this graph is that even at 4 kHz operation, the faster device still offers lower losses at reduced current levels.  This can be explained by the fact that conduction losses have a stronger dependence on current than switching losses.  A first order model of conduction losses is proportional to the square of the current while the switching losses are proportional directly to current. 

Further considerations for achieving higher RMS current output and higher efficiency is to use Solderable Front Metal (SFM) IGBTs and IGBTs with lower Short Circuit time (TSC).

SFM IGBTs enable solderable emitter interconnects and hence reduce interconnect impedance; reduce IGBT junction temperature through enabling advanced cooling methods. Reduced junction temperatures reduce VCEON & ETS of the IGBTs and hence improve the Inverter performance.

Reducing the minimum short circuit time requirements from 10 uS to about 6 uS can improve the RMS current output and the efficiency in inverter systems. Reduced TSC is achieved by increasing the gain of the IGBT, which results in lower VCEON & ETS in the application. Reducing VCEON & ETS improves the system performance. The modern drive and protection circuits have smarter noise filtering capabilities and have faster response times than their predecessors and hence can protect the IGBTs within 6 uS. Hence, reducing the TSC requirement increases output RMS current and efficiency of the inverter.

Matching the VCEON and EOFF trade-off in the application conditions is critical to achieving required output RMS current and highest efficiency of a motor drive system.


References:
1. International Rectifier Applications Note AN-983 IGBT Characteristics
2. International Rectifier Applications Note AN-990 Application Characterization of IGBTs 

Advertisement

Share this Story

X
You may login with either your assigned username or your e-mail address.
The password field is case sensitive.
Loading