This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LMX2491: noise issue on CPOUT

Part Number: LMX2491
Other Parts Discussed in Thread: PLLATINUMSIM-SW

Dear All,

 I have a noise issue on CPOUT(exactly, VCO after loop filter) of LMX2491.

This noise shows on VCO node after loop filter like below waveform.

I think that there are some problems on loop filter or register setting.

Please review the below schematic/waveform/register setting, and then share your advice with me.

1. Schematic

1663.LMX2491.pdf

2. Waveform

3. Register setting

MClk100Mhz_DetectPhase50MHz_FMCW_ReverseSawtooth_Band198MHz_1ms.txt
R141	0x008D00
R140	0x008C00
R139	0x008B32
R138	0x008A00
R137	0x008900
R136	0x0088A3
R135	0x008798
R134	0x0086E0
R133	0x0085C3
R132	0x008450
R131	0x008300
R130	0x008200
R129	0x008100
R128	0x008000
R127	0x007FC0
R126	0x007EC3
R125	0x007D50
R124	0x007C00
R123	0x007B00
R122	0x007A00
R121	0x007900
R120	0x0078A0
R119	0x0077C3
R118	0x007650
R117	0x007500
R116	0x007400
R115	0x007300
R114	0x007200
R113	0x007120
R112	0x007000
R111	0x006F32
R110	0x006E00
R109	0x006D01
R108	0x006C44
R107	0x006B38
R106	0x006A61
R105	0x006961
R104	0x0068A8
R103	0x0067BF
R102	0x0066FF
R101	0x0065FF
R100	0x00645A
R99	0x006348
R98	0x006200
R97	0x006132
R96	0x006000
R95	0x005F00
R94	0x005E00
R93	0x005D00
R92	0x005C24
R91	0x005B00
R90	0x005A32
R89	0x005900
R88	0x005801
R87	0x005799
R86	0x0056FE
R85	0x005500
R84	0x005400
R83	0x005300
R82	0x005200
R81	0x005150
R80	0x005013
R79	0x004FA9
R78	0x004E00
R77	0x004D10
R76	0x004CA3
R75	0x004BD7
R74	0x004A00
R73	0x004900
R72	0x004800
R71	0x004700
R70	0x004603
R69	0x004500
R68	0x0044E2
R67	0x004300
R66	0x004200
R65	0x004100
R64	0x004000
R63	0x003FE2
R62	0x003E00
R61	0x003D00
R60	0x003C00
R59	0x003B09
R58	0x003A11
R57	0x003900
R45	0x002D00
R44	0x002C00
R43	0x002B00
R42	0x002A00
R41	0x002900
R40	0x002800
R39	0x0027F2
R38	0x0026FA
R37	0x002517
R36	0x00240F
R35	0x002341
R34	0x002204
R33	0x002120
R32	0x002000
R31	0x001F32
R30	0x001E0A
R29	0x001D00
R28	0x001C1F
R27	0x001B08
R26	0x001A00
R25	0x001902
R24	0x001800
R23	0x001700
R22	0x001600
R21	0x001500
R20	0x001400
R19	0x001300
R18	0x00122C
R17	0x001100
R16	0x00101E
R15	0x000F00
R14	0x000E00
R13	0x000D00
R12	0x000C00
R11	0x000B00
R10	0x000A00
R9	0x000900
R8	0x000800
R7	0x000700
R6	0x000600
R5	0x000500
R4	0x000400
R3	0x000300
R2	0x000201
R1	0x000100
R0	0x000018

Regards,

Jeffrey

  • Jeffrey,

    Here's my comments:

    1. Schematic:
    On the Fin/Fin* pins you have 10 pF AC coupling cap. Consider increasing this. Sometimes fear of self-resonant frequency causes people to put a very small cap here but then what happens is the impedance is too high and then we run into issues with sensitivity, or higher harmonics of the VCO passing through the coupling cap eaiser than the fundamental and causing issues.

    Also, on Fin#, you might just use 47 or 51 ohms and not 68 ohms, although this probably won't make a huge difference.

    It's hard to comment on the loop filter as this also depends on VCO gain, charge pump gain, VCO frequency, phase detector frequency , and potentially VCO input capacitance if it is significant compared the 470 pF you have. Actually, from the register file, it seems phase detector requency is 50 MHz, output frequency is 1.5 GHz, charge pump gain is 3.1 mA. So what is missing is VCO gain and VCO input capacitance. If you know these values, you can model your loop filter in the PLLatinum sim tool (ti.com/tool/PLLATINUMSIM-SW)

    2. Waveform
    The purple trace seems after the loop filter and maybe the light blue is before? If so, my first instinct would be to short R50 and see if the issue goes away. If so, maybe the VCO has high input capacitance.

    3. Registers
    I was able to load these in TICSPro, but unfortunately the Ramp GUI tab does not update properly from just a register load; it likes the saved TICSPro file. So I can't comment on the ramp tab, but the other part of the setup looked reasonable to me.

    Regards,
    Dean
  • Hi Dean,

    Thanks for your comments.
    Let me check your comments like above, then I'll feedback youl

    Regards,
    Jeffrey