Non Invasive Stability Measurements vs. Bode Plots

Size: px

Start display at page:

Download "Non Invasive Stability Measurements vs. Bode Plots"

Steven O’Connor’
5 years ago
Views:

1 Comparisons of Non Invasive Stability Measurements vs. Bode Plots AEi Systems 5777 W. Century Blvd. Suite 285, Los Angeles, CA info@aeng.com More Information: invasive stability measurement.html 24 AEi Systems. All Rights Reserved

2 What s Included Here Hardware circuitry including Linear Regulators, POLs, and Switchers were constructed and bench tested Non Invasive Stability Measurement ( ) of Stability Margins and Bode Plots were both recorded dd In some cases, the circuit were also simulated TheStability Margin vs. Phase Margin results are compared and documented Note: The testing was performed by AEi Systems. 24 AEi Systems. All Rights Reserved 2

3 What is and Why It s a Critical ii Technology Non Invasive Stability Measurement is a method of determining control loop stability margins without requiring access to the feedback loop In many situations it is not possible to access the control loop Examples POLs, Fixed Voltage Regulators, Voltage References, High BW Opamps Integrated Switching ICs that do not allow loop access In other cases it might be impractical to break the control loop because cutting a printed circuit board traceorlifting components might be required is computed by converting output impedance to group delay. Then the Q is derived from the group delay and the stability margin from the Q The technology is licensed and promoted by Picotest.com The measurement capability can be found on various VNAs OMICRON Lab, Keysight, Copper Mountain, Rhode Schwarz 24 AEi Systems. All Rights Reserved 3

4 is Based on Fundamentals Dr. R.D. Middlebrook popularized the topic of Minor Loop Gain, T m with his introduction of the extra element theorem which allowed us to assess the stability of power supplies and input filters Minor loop gain, based on Nyquist criteria is now one of the most researched electronics topics Many articles can be found with an internet search of forbidden region stability criteria The concept is simple. Break a system into two parts, generally termed a System and a Load and determine the impedance of each part, ZS and ZL Z s Z L Phase margin is determined by setting Tm = and solving for phase 24 AEi Systems. All Rights Reserved 4

5 is Based on Fundamentals Z s Z L Acomplication exists in that we cannot separate the voltage regulator into two parts, Z S and Z L without cutting a trace orremoving removing the capacitor software extracts data from impedance and group delay and allows the Z S and Z L to be mathematically separated so they can be converted to Stability Margin Practically speaking, impedance is measured with a suitable probe in a or 2 port configuration The instrument converts the impedance to group delay and Q The user positions waveform cursors on the impedance and Q waveforms and the conversion to phase margin is read out on the instrument s screen A video of the process can be viewed here, software.htmlpicotest com/products software html 24 AEi Systems. All Rights Reserved 5

6 Non Invasive Stability Measurement vs. Bode Plot Measurement RH86 Linear Regulator RH85 Linear Regulator LM37 Linear Regulator Various Configurations TPS422 Buck Regulator TPS7A45 Linear Regulator LMR55 Simple Switcher TL43 Adjustable Shunt Regulator VRG8666 (RH38) Linear Regulator CLC7 245MHz Opamp 24 AEi Systems. All Rights Reserved 6

7 RH86 Linear Regulator Bode Plot PM Measured 59 deg 56deg Simulated 63 deg 56 deg 5.V Vin IN X2 RH86 OUT.8V Vout Vout Min = 25mA Nom = 5mA Max = A Vin 5 +/-.5V Cin T49G226M5 ADJUST 557mV R % C C2.uF.uF 45.24% 45.24% I 25m Cout ADJ freq Vf req Cadj V R2 68n 3.28% 5.87% R3 B Current mag(f req)/(2* ) I2 AC = 24 AEi Systems. All Rights Reserved 7

8 RH86 Linear Regulator 2 2 db_v(vout) 5 tgqcurve NIP = 63 degrees.8. x = 9.7k hertz, y = -2.9 db(volts) TR - -2 TR PM: Cursor 52.58k58k m 54.29m Cursor k 32.57m m C2-C -.83k -22.6m 93.28m TR: Mag(Gain) : QTg(Gain) - Plot tgqcurve.4. 6m 2m b_v(vout) in db(volts) d x = 57.9k hertz, y = 456m k k k Meg Meg frequency in hertz 2 5 /db TR 5-5 TR/dB / Cursor k : Phase(Gain) - 2/ ot ot) in db(volts) Plo db_v(bodeplo ot) in degrees ph_v(bodeplo ph_v(bodeplot) 2 db_v(bodeplot) BW = 73.4k hertz, PM = 56.2 degrees k k k Meg Meg frequency in hertz Bode Plot PM Measured 59 deg 56 deg 2 24 AEi Systems. All Rights Reserved Simulated 63 deg 56 deg 8

9 RH85 Linear Regulator 2 TR TR PM: -2 Cursor k m m > 7 Cursor k m m C2-C C k m m TR: Mag(Gain) : QTg(Gain) 3 Bode Plot PM Measured > 7 deg 94 deg TR /db 5-5 TR/dB / Cursor k : Phase(Gain) - TR R2/ Vin V 5V +/-.25V IN X4 RH85 ADJUST C TBJD336KC OUT Adj C7 2p 5.87% freq R5 Vout R % R % Vfreq Vout C2 TBJD336KC8 R3 R4 6.28% 6.28% 5 B Current mag(freq)/(2* ) C3 TBJD336KC C6.uF 45.24% I2 AC = C5.uF 45.24% C4.uF 45.24% Iload 25m 24 AEi Systems. All Rights Reserved 9

10 LM37 Linear Regulator, Cap 2 TR/dB - -2 TR/dB PM: Cursor 8.8k Cursor k C2-C m m : QTg(Gain) - -2 Bode Plot Measured 24 deg 25 deg TR/dB TR/dB / Cursor 7.854k Memory : Mag(Gain) Memory : Unwrapped Phase(Gain) / General LM37 configuration used, Cap is a uf Tantalum Note: The circuit is from the Picotest VRTS board, It was measured using several capacitors in the kit. 24 AEi Systems. All Rights Reserved

11 LM37 Linear Regulator Cap 4 2 2* TR/dB 5 8* - 6* - 4* - -5 TR/dB PM: 2* - Cursor k Cursor k C2-C k m 3.597m * 4 4* 4 6* 4 8* 4 5 : QTg(Gain) Bode Plot R/dB T 5-5 TR/dB / Cursor k f : Phase(Gain) 5-5 / Measured 29 deg 3 deg Measured on the VTRS board, Cap 4 is a 2.2uF Tantalum 24 AEi Systems. All Rights Reserved

12 LM37 Linear Regulator Cap TR/dB TR/dB PM: Cursor k m Cursor k m C2-C -3.3k m m : QTg(Gain) -2 Bode Plot 2 Measured 56 deg 53 deg 5 R/dB T -5 / Measured on the VTRS board, Cap 5 is a 5uF Tantalum TR/dB / Cursor 26.83k : Phase(Gain) 24 AEi Systems. All Rights Reserved 2

13 LM37 Linear Regulator Cap 6 2 TR/dB TR/dB PM: -5 Cursor 27.86k Cursor k C2-C m m 4 2* 4 4* 4 6* 4 8* 4 5 : QTg(Gain) - Bode Plot Measured 9 deg 8 deg 5 5 TR/dB -5 TR/dB / Cursor k f Memory : Mag(Gain) : Phase(Gain) Memory : Phase(Gain) -5 / Measured on the VTRS board, Cap 4 is a 22uF Ceramic 24 AEi Systems. All Rights Reserved 3

14 TPS4222 Buck Regulator TR/dB TR R/dB TR/dB PM: -4-2 Cursor 8.424k Cursor k C2-C 3.99k m m : QTg(Gain) 3.8 deg khz TR/dB / Cursor 99.72k : Unwrapped Phase(Gain) / Bode Plot Measured 3 deg 32 deg A 5 V Input, 6A.6 A Output, Non Synchronous Buck Converter 24 AEi Systems. All Rights Reserved 4

15 TPS7A45 Linear Regulator ma TR - TR R/dB TR PM: Cursor.857k m 6.98 Cursor k m C2-C m 29.48m ma : Mag(Gain) ma : QTg(Gain) TR/dB / Cursor.775k 4.2f ma : Mag(Gain) ma : Phase(Gain) / Bode Plot Measured 6 deg 54 deg A low noise, fast transient response 5A.5 A lowdropout (LDO) voltage regulator 24 AEi Systems. All Rights Reserved 5

16 TPS7A45 Linear Regulator ma TR - TR R/dB TR PM: Cursor 3.366k m 47.7 Cursor k C2-C m 42.8m ma : Mag(Gain) ma : QTg(Gain) TR/dB / Cursor 3.43k 42.53m ma (k) : Mag(Gain) ma (k) : Phase(Gain) / Bode Plot Measured 47 deg 47 deg A low noise, fast transient response 5A.5 A lowdropout (LDO) voltage regulator 24 AEi Systems. All Rights Reserved 6

17 TPS7A45 Linear Regulator 5mA TR - TR R/dB TR PM: Cursor 6.65k m 5.24 Cursor k m C2-C m 24.74m mA : Mag(Gain) 5mA : QTg(Gain) TR/dB / Cursor 7.53k 45.24m mA (2.37k) : Mag(Gain) 5mA (2.37k) : Phase(Gain) / Bode Plot Measured 5 deg 53 deg A low noise, fast transient response 5A.5 A lowdropout (LDO) voltage regulator 24 AEi Systems. All Rights Reserved 7

18 LMR55 Simple Switcher POL TR/Ohm TR/dB 4* - 2* - - 8* -2 6* -2 2 TR/Ohm PM: 4* -2 Cursor 59.28k m 9.33m Cursor k m 7.7 C2-C.33k m 6.52m 2* TR: Mag(Impedance) : QTg(Impedance) TR/dB / Cursor 46.29k.926m : Phase(Gain) / Bode Plot Measured 5 deg 5 deg A 5V to 33PPOL 3.3P POL (Simple Switcher Step Down Buck Regulator) AEi Systems. All Rights Reserved 8

19 VRTSP5 TL43 Adj. Shunt Regulator 6 2 TR/dB TR/dB PM: -4 Cursor 6.53k Cursor k C2-C m 93.97µ : QTg(Gain) Bode Plot Measured 36 deg 35 deg R/dB T TR/dB / Cursor 6.535k : Phase(Gain) / An Adjustable Shunt Regulator with 2.5V Reference / VRTS3h 24 AEi Systems. All Rights Reserved 9

20 VRG8666 (RH38) Linear Regulator 2 - TR/dB TR/dB PM: -2-4 Cursor 6.67k Cursor k C2-C m 73.76m : QTg(Gain) Bode Plot Measured deg deg 8 2 TR/dB TR/dB / Cursor 6.652k f : Phase(Gain) / V 5 C2 33u 3 Vc Vin X VRG8666 Inj SET 2 Bode R 25k Vout 4 C u 2.5V C3 33u I 25m 24 AEi Systems. All Rights Reserved 2

21 VRG8666 (RH38) Linear Regulator 2 TR/dB TR/dB PM: Cursor 4.88k Cursor k C2-C m 5.7m : QTg(Gain) Bode Plot Measured 2 deg 2 deg TR/dB TR/dB / Cursor 4.33k : Phase(Gain) / V 5 C2 33u 3 Vc Vin X VRG8666 Inj SET 2 Bode Vout R 25k 4 C u 2.5V C3 33u I 25m R AEi Systems. All Rights Reserved 2

This is a great capability; to be able to accurately assess stability at 's of MHz or higher without lifting any wires (which would interfere with the measurement).

22 CLC7 245MHz Opamp Using the Keysight E56B VNA, with the software, we were able to test the stability of this 245 MHz opamp. The big WOW is that we obtained the (very poor) phase margin from the impedance measurement using (just about 2 degrees). This is a great capability; to be able to accurately assess stability at 's of MHz or higher without lifting any wires (which would interfere with the measurement). While there is no loop correlation for this test, The measured phase margin of 2.36 is not acceptable because signals near the 63MHz bandwidth would become very distorted and fast transitions would see oscillations at this high frequency. Bode Plot Measured 2 deg N/A 245 MHz opamp 24 AEi Systems. All Rights Reserved 22

CBC2 performance with switched capacitor DC-DC converter. systems meeting, 12/2/14

CBC2 performance with switched capacitor DC-DC converter systems meeting, 12/2/14 1 reminder of results from CBC1 +2.5V GND 1n can power CBC from single +2.5V supply 1uF DC-DC diff. clock (CMOS) 1 MHz