Article count:230 Read by:1375152

Account Entry

Case study: Problems with using MOS as a switch in the total voltage sampling circuit

Latest update time:2022-04-01
    Reads:

After being isolated at home for a period of time, I finally went to work. I am still not used to working from home. I feel that the effect of work and rest is not very good. I am tired from work and don’t get enough sleep.

Let me share a case today, which is about some problems when using high-voltage MOS as a voltage dividing resistor switch.
A previous article " PhotoMOS and Reed Switch--- Talk about Isolation Devices in Insulation Detection Circuits " mentioned the use of high-voltage MOS as a switch. For example, in the typical application circuit of LTC2949 shown below (from the ADI official website), high-voltage MOS is used as the bridge arm switch for insulation detection; the reason for choosing high-voltage MOS is that the cost is relatively lower than that of optical MOS.

The common principle diagram of using high-voltage MOS as a switch in total voltage sampling is as follows: R1 and R2 are voltage dividing resistors, and the ADC calculates the total voltage by collecting the voltage on R1; the position of NMOS is different from the general usage, and its S level It is not directly connected to ground, but connected to the sampling resistor R1; R3 and R4 are the driving voltage dividing resistors of NMOS.

Question 1: Selection of NMOS driving voltage
Take ST's high-voltage MOS (model number STD3N95K5AG) as an example, as shown below (from ST's official website): Its VDS can withstand a voltage of up to 950V.

When we choose MOS as a switch, we usually check its driving threshold voltage, as shown below: VGS(th) is between 3V~5V, so the driving voltage must be >5V.

In addition, in order to minimize the resistance when it is turned on and avoid causing sampling errors, we hope that the MOS tube is in a saturated conduction state, so its driving voltage is preferably greater than the Miller platform voltage, which is about 8.5V. (There is a lot of information about the Miller Platform. To be honest, I don’t have a complete grasp of it)

In addition, there is another place that is easily overlooked, which is the voltage division on the sampling resistor R1. If its maximum voltage division in the full voltage range is 5V, then the MOS driving voltage must also include this part, that is, the minimum The driving voltage is 8.5V+5V=13.5V.

Question 2: IGSS current problem
MOS tubes are different from transistors. They are voltage-driven, but this does not mean that their G and S terminals do not need to consume current. It is just that this has basically no effect when MOS is used as a digital switch; but in today's application, IGS cannot Ignored, the figure below shows the maximum IGSS of this MOS: ±10uA, which is actually very large.

Specifically, as shown in the figure below: This current IGS will also flow through the sampling resistor R1, which will inevitably cause a sampling error on R1. Especially at high temperatures, IGS will increase, causing the sampling error to also increase; this error is It is unavoidable and can only be reduced as much as possible. One solution is to reduce the VGS driving voltage of NMOS and adjust it to a voltage point that can fully turn on the MOS and meet the accuracy requirements .

Question three: IDSS current problem
This refers to the leakage current at both ends of DS when the MOS tube is not turned on (VGS=0); the IDSS of this model of MOS is shown in the figure below, and the leakage current can reach 50uA at high temperatures.

Its impact is that when we do not drive the MOS tube to turn on, due to the existence of IDS leakage current, there is a voltage drop on R1. At this time, a voltage will be collected, but in theory the voltage should be 0V at this time, so Contradictions occur, and in severe cases, they may cause misjudgments. For example, when the voltage sampling point at the rear end of the relay is not closed, the voltage should not be collected.

Summarize:
That’s all for now. Tomb-Sweeping Day is coming soon, and the current situation is that I can’t go anywhere and just stay at home; all the above are for reference only.

Latest articles about

 
EEWorld WeChat Subscription

 
EEWorld WeChat Service Number

 
AutoDevelopers

About Us Customer Service Contact Information Datasheet Sitemap LatestNews

Room 1530, Zhongguancun MOOC Times Building,Block B, 18 Zhongguancun Street, Haidian District,Beijing, China Tel:(010)82350740 Postcode:100190

Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved 京ICP证060456号 京ICP备10001474号-1 电信业务审批[2006]字第258号函 京公网安备 11010802033920号