SET Curve Tracer Dan's Questions and Notes ---------------------------------------------- Original Rev. 25-Nov-2019 Current Rev. 3-Sep-2020 1. Quick Review of: Instrumentation Amps, Op-Amps, Input Bias current, Johnson Noise, Input Noise Voltage, Input Noise Current, Total Noise Input, expected S/N ratio 2. Verify understanding of the setup used so far, layout with all low level inside the cryostat, move instrumentation amps inside the cryostat 3. Use a stand alone test setup, i.e. not the cryostat, for a first test and to set a baseline of how well it could work. 4. Other microwave RF components: RF Ground connection to SET, Why a Bias-T ?, RFC, Y/4 Choke, Y/4 or Y/2 line to match to Hi-Z, Cavity to match to Hi-Z, minimize loss to have best chance to see delta loss in SET 5. Gate connection to SET ? If the box we are making already provides the Source-Drain voltage and the Offset from Ground voltage should it also provide the Gate voltage ? 6. RF Filter items to check: direction, micro-phonics, low frequency peak, 7. Instrumentation Amp pcb current size 31.0 mm x 19 mm. 8. Are there amplifiers better than the INA110 ? INA110 OPA827 ------- ------- Bias Current 25C typ 20 * 3 pA Bias Current 25C max 100 * 10 pA Bias Current 90C aprox 1000 500 pA * 2x less Bias Current on selected grade of INA110. Bias Current vs Comm Mode Voltage Noise at 0.1 Hz - 60 nV/root Hz typ Voltage Noise at 1 Hz 110 19 Voltage Noise at 10 Hz 40 7 Voltage Noise at 100 Hz 15 4.5 Voltage Noise at 1k Hz 10 4 Voltage Noise at 10k Hz 10 3.8 Noise 0.1 Hz to 10 Hz 1 0.25 uVpp typ Noise Current 1.8 2.2 fA/root Hz at 10k Hz for the INA110 at 1k Hz for the OPA827 Initial OffSet Voltage typ 100 * 75 uV Initial OffSet Voltage max 500 * 150 OffSet Volt vs Temp typ 2 * 0.1 uV/Deg C OffSet Volt vs Temp max 5 * 2 * 2x less OffSet Voltage and Drift is available on selected grade of INA110. Power Supply Current typ 3.0 4.8 mA Date Introduced about 1985 about 2006 8. Power supply for amplifiers have the normal issues of noise and ground loops. Must understand grounding system for cryostat. 10. Specific Proposal: - Build an ideal test setup for DC measurement of the SET that does not involve the cryostat at all for now. - Make serious measurement of the noise of this test setup in the low frequency aka DC range. - Note that this circuit is basically self testing, e.g. with test resistors installed instead of a SET then increasing the Source to Drain supply voltage by 10% should increase the channel voltage and current readouts by 10%. ================================================== Other Questions, Points, and Descriptions: ------------------------------------------ Rev. Date 6-Feb-2020 1. List of the required Circuit Boards: Instrumentation Amplifier Signal Filters 1 Hz and 100 Hz use same pcb Power and Bias Feedthrough Filters 25 kHz Power Supply 2. Drawing List: Overall Physical Block View Stuff Inside the Cryo Stuff On Top of Cryo Power Supply Block Diagrams Power Supply Schematics Power & Bias Filter Assembly on top of the Flange Power & Bias Filter Box Schematics Signal Filter Box Schematics 3. Details of the Internal Heat Load: The heat fill all be from the quiescent current in the opamps that make up the two instrument amplifiers. Still not clear whether to use OPA827 or OPA828. Both the 827 and 828 are rated for operation over -40 to +125 deg C and from +-4V to +-18V power supply. OPA827 quiescent current: 4.8 mA typ 6.0 mA abso max. OPA828 quiescent current: 5.5 mA typ 7.9 mA abso max. OPA827 figure 16 shows that its quiescent current goes monotonically down as a function of temperature. OPA828 figure 15 shows that its quiescent current goes monotonically down as a function of temperature. Lets assume that we will operate the internal amplifiers from +-10V supplies, i.e. right in the middle of their operating range. Note that things like Bias Current and Offset Voltage (and the drift of both of these) look fine for +-10V operation. A higher supply voltage will be necessary if the SET Bias Offset is much above +- 3 Volts. Operating inside the top of the cold cryostat there is little chance that the quiescent current will be over the typical value. Thus, per opamp the heat should be about 110 mW. For all 6 inside opamps this is 660 mW total. 4. Where How to make the Ground Connections to the Cryostat: ??? The intent is to have a common System Ground Point on the top head of the cryostat and to use individual "star" ground connections from there. That is, the ground currents from item A do not flow through the ground connection for item B. Can the Flange Cover Plate over the Instrument Amplifiers be used as the Ground Point, i.e. do the copper gasket and bolts holding the cover plate on make a good enough connection to the rest of the cryostat ??? Number of External Ground Connections Needed to the Cryostat Ground Point: Instrument Amp Power Filter First Stage & Return \ Common Instrument Amp Power Filter Second Stage / V & I SET Bias Supply Filter First Stage \ Common SET Bias Supply Filter Second Stage / Value and Offset SET Bias Offset Supply Return SET Bias Supplies Case Ground Signal Ground for the Voltage Signal Filter Input Signal Ground for the Voltage Signal Filter Output Signal Ground for the Current Signal Filter Input Signal Ground for the Current Signal Filter Output Number of Internal Ground Connections Needed to the Cryostat Ground Point: V Ch Inst Amp Gnd Plane V Ch Inst Amp Reference I Ch Inst Amp Gnd Plane I Ch Inst Amp Reference Internal Shields ??? 5. Capacitor and Feedthrough Capacitor Examples: 10 nFd 15.0 mm lead space 18 x 5 mm footprint R75RI21004030J 47 nFd 15.0 mm lead space 18 x 5 mm footprint R75PI24704030J 100 nFd 15.0 mm lead space 18 x 5 mm footprint R75MI31004030J 1 uFd 15.0 mm lead space, 18 x 10 mm footprint R75GI41004000J 4.7 uFd 27.5 mm lead space, 32 x 13 mm footprint R75GR44704030J 10 uFd 27.5 mm lead space, 32 x 22 mm footprint R75GR51004000J These are all: Polypropylene, +- 5%, non-inductive, about -200 ppm/deg C, 30k Meg_Ohm x uFd min 150k typ, Kemet R75 series, radial, 160 V DC (except for 100, 47, 10 nFd which are higher voltage to get the same case size) Some capacitor foot-prints on the Signal Filter pcb should be setup for two lead spacings, i.e. 15.0mm and 27.5mm. All resistors are in the 1k Ohm to under 100k Ohm range. An issue may be THD vs SMD. Even at big 1206 the SMD look more available than the THD - 1% and 100 ppm looks standard - 100 ppm / deg C takes 100 deg C to get 1%. This is OK for the external Filters and such. Bulk Power Filter Caps: Polymer-Al Electrolytic: under/about 1.5 mAmp leak >> 330 uFd 25 V 14 mOhm 105 deg C 5000 hrs 25SEPF330M 5.0 mm Lead Spacing 0.6 mm Lead Diameter 10.0 mm Diameter 13.0 mm Height Negative Polarity Mark 470 uFd 25 V 14 mOhm 105 deg C 1000 hrs 25SEK470M Al Electrolytic: under/about 350 uAmp leak 1000 uFd 35 V 18 mOhm 105 deg C 5000 hrs ESY108M035AL4AA 1300 uFd 35 V 30 mOhm 105 deg C 5000 hrs EEU-FP1V132 >> 1000 uFd 35 V 38 mOhm 105 deg C 5000 hrs EEU-FP1V102 5.0 mm Lead Spacing 0.6 mm Lead Diameter 12.5 mm Diameter 22.0 mm Height Negative Polarity Mark SET Bias Filter Caps: same as signal filter caps and need a larger value cap 100 uFd 37.5 mm lead space, 41.5 x 20 mm footprint R60EW61005000K Kemet R60 Polyester R60EW61005000K 100 uFd 100 Volt radial thd 41.5 mm x 20 mm 40.1 mm tall 37.5 mm Lead Spacing 105 deg C Polyester, Polyethylene Terephthalate (PET), Metallized 10% tolerance cost about $15 leaks 1.25k Meg Ohm x uFd max leaks 5k Meg Ohm x uFd typical Look at the Tantalum capacitors: T491, T494, T495, T510 in the 100 uFd 20 V these all leak 20 uAmp T489 (low special leakage) 100 uFd 20 V leaks 15 uAmp Look at the Polymer Tantalum capacitors: T521, T523, T541, T543, T598 100 uFd 20 V all leak 200 uAmp The wet Tantalum in hermetic package 100 uFd 15 V or 20 V leak only 6 uAmp to 10 uAmp see Sprague 150D or Kemet T110 or Kemet T212. But all Tants and Electrolytic require a mono-polar SET Bias Offest. Look at Niobium Oxide 100 uFd 10 V leaks 20 uAmp Look at all kinds of Al-Electrolytic caps 1000 uFd 35 V basically everything leaks 350 uAmp - special stuff Vishay BComponents 013 RLC (or 148RUS) may only leak 70 UAmp, i.e. 0.002 x C uFd x Volt Rating instead of 0.01 time the uFd Volts with leakage in uAmps. All Polymer-Al caps had much higher leakage than the plan old Al-Electrolytic caps. Bulk Polymer Tantalum capacitors for bypass in the Signal Filter Box could be 100 uFd 25 Volt 30 mOhm 2917 / 7343 Kemet T521X107M025ATE030 Bulk ceramic for bypass in the Signal Filter Box 10 uFd 25 V 1206 Kemet C1206C106K3RACTU Ceramic bypass for use in the Signal Filter Box 100 nFd 50 Volt X7R 0805 Kemet C0805C104M5RACTU NP0 / C0G Bypass Capacitors for the Instrumentation Amps 100 nFd 25 V NPO 1206 C1206C104J3GACAUTO and C1206X104K3GECAUTO 47 nFd 25 V NP0 0805 C0805C473K3GAC7800 same but automotive C0805C473J3GACAUTO 22 nFd 50 V NP0 0805 Automotive C0805C223J5GACAUTO 10 nFd 50 V NPO 0805 Automotive C0805C103K5GECAUTO RF Feedthrough Capacitor example: 100 pFd and other about $8 each Manufacturer Tusonix a Subsidiary of CTS Electronic Components Manufacturer Part Number 2425-018-X7R0-101MLF 6. CF (ConFlat) Flanges In North America, flange sizes are given by flange outer diameter in inches, while in Europe and Asia, sizes are given by tube inner diameter in millimeters. Despite the different naming conventions, the actual flanges are the same. European, Asian North American size --------------- ------------------- DN10 1" DN16 1 1/3" ("mini") DN25 2 1/8" DN40 (or DN35) 2 3/4" DN50 3 3/8" DN63 4 1/2" DN75 4 5/8" 7. Signal Filter Box - Details: Does the Signal Filter Box need: - a 10:1 attenuator - series parallel selection for filter banks - buffer for the output after the first series filter Power supply design for the Signal Filter pcb: level of regulation, noise filter. Signal Filter Box: size, mounting, controls, indicators, connectors, 8. Power Supply for the Instrumentation Amplifiers: The Instrument Amps will use +- 10 Volt supplies. The expected load of the 6 OpAmps is 33 mA. This is 660 mW into the cryostat. Capacitor selection as described above in item 5. Pass-Transistors assume Beta => 100, expect to dump about 3.5 V x 35 mA = 123 mW. Need to select/verify low noise of pass transistors at low frequency. 25 kHz is 40 usec period. 33 mA will charge 1.32 uFd to 1 Volt in 40 usec. 1 Volt would be 10% ripple. Rectification will be full wave so this is cut in half. 330 uFd is 250x 1.32 uFd. Leakage of this size Al-Polymer capacitor is about 1.5 mA. 1000 uFd 35 V Al-Electrolytic is about 350 uA. Set the transformer secondary turns to give about 15 Volt bulk supplies with choke input filter. The Base Zener is set for 11 Volts. The R1 drop resistor from bulk to the zener carries 5 mA zener current, 350 uA base current, and 350 uA C2 Al-Elec filter cap current and has a 4 Volt drip. Thus the nominal size of R1 is 701 Ohms. The Collector Resistor R2 should drop about 1.5 Volts with the 35 mA of collector current. Thus the nominal size of R2 is 43 Ohms. The Zener to Base Resistor R3 carries about 350 uA of Base current and about 350 uA of Al-Elec C2 filter cap current and drops about 1 Volt. Thus the nominal size of R3 is 1429 Ohms. The output protection zener is set for 14 or 15 Volts. If the pass transistor shorts then this 500 mW zener will force any required drop across the R2 collector resistor. Need to use separate noisy and quiet ground connections from the power_supply/filter to the cryostat. Need to verify that the OpAmps will be happy with the closest big caps being the 330 uFd Al-Polymers at the output of this supply, i.e. the other side of the vacuum feedthrough. Do we need some small Tants on the Instrument Amplifier pcb ??? 9. Toroid Transformer and Filter Inductor Designs: For the 3E2A 0.870" OD cores the initial inductance is about 3055 mH per 1000 turns and drops to about 670 mH at 3000 Gauss, so 10 turns is about 0.31 mH with no DC current. For a 20 mA "magnetization" current in the primary of a 25 Vpp 25 kHz sin wave transformer made with two of the 0.87" OD 3E2A cores I should use about 17 turns. The effective load on one of the 15V Instrument Amplifier power supplies is about 15V / 33mA = 455 Ohm. For a current ripple of 10% we want .... Recall that 1 mH at 25 kHz is 157 Ohm reactance. Need to calculate the Amp-Turns to H magnetization force for both size cores to verify that they are well out of saturation. The 3E2A 0.870" OD cores saturate at about 2.6 Apm-Turns. 10. For the low frequency low noise DC stable Instrumentation Amplifiers the OPA827 may be a little quieter than the OPA828 but the OPA828 has lower bias current so it may win by having less current noise. The ADA4625 may be just as good or better. Would need to test at the operating temperature to learn which is best. They are probably all bias current compensated and who knowns what that does at -55 deg C. If one wants to try a commutating amplifier then the OPA189 looks interesting but has higher bias current and probably lots of generate noise in its bias current. 11. An good temperature sensor to use may be the inexpensive TDK type B57330V2103F260. This is 10k Ohm at 25 deg C and goes up to about 446k Ohm at -55 deg C. The calibration at 25 deg C is 1%. This is a solder on 0603 SMD part. 12. The SET Bias power supplies are: Yokogana model GS200. ================================================== Notes from the Meeting with Niyaz on Thursday 6-Feb-2020: --------------------------------------------------------- Rev. Date 7-Feb-2020 1. The flanges on the top of their cryostat are type KF (not CF). KFs are the ones with a separate center O-ring section and a clamp that goes around the outside. I don't think that one can count on them making a good stiff electrical connection using a KF flange (the O-ring is insulating and most of the clamps are die-cast aluminum and thus make a poor unstable electrical connection). The current day-dream is that the SET Bias Filter and the Instruentation Amp Power Supply and its Filter will all need to go in a box that is about 1 foot from the KF flange and this box will need to connect to the Fischer Connector on the KF Flange via a cable. I mentioned also running a serious ground braid cable to a hose clamp on the pipe stub going to the KF flange. 2. The ID of the pipe where the Instrumentation Amps can go is 40 mm. 3. The hermetic electrical connectors that they like are from a company named Fisher. I think that these connectors are from the "brass core" series. 4. The vacuum in the area where the Instrumentation Amps go is 10e-5 mbar. Recall: Torr == 1/760 standard atmosphere Torr = very very very close to 1 mm of Hg Torr is about 0,019 psi Torr is not an SI unit bar == 100 kPa bar is about 0.986 of a standard atmosphere so 1 Torr is about 1.3 mbar 5. Yes, the SET Bias Offset needs to be BiPolar. 6. They may want to change the SET Bias Voltage at up to 100 times per second. It was not clear to me that all operation of the SET transistor needed the SET Bias to change at 100 Hz or just some specialised modes of operation (e.g. lockin) will need this high rate. Thus it may be necessary to have two different corner frequencies for the Set Bias Filter. The SET Bias Offset will only need to be changed much more slowly, e.g. at 1 Hz. This implies that the SET Bias Filter could have a low corner frequency on its Common Mode but will need a higher or selectable Normal Mode corner frequency. 7. The SET Bias power supplies are Yokogana model GS200. I believe that the GS200 supply has a Guard connection. Wiring of the remotely located SET Bias power supplies is an issue. Recall that the Bias Voltage supply has both of its outputs floating. Their initial plan was to use BNC cables. There is also the issue of making the connection between the Voltage and Offset supplies at the supplies or in the SET Bias Filter box. 8. The nominal resistance of the SET is probably closer to 40k Ohm than the original estimate of 100k Ohm. We probably want to move the current shunt from the original sketch of 300k Ohm down to 100k Ohm. 9. Kurt Lesker company seems to be the easy source KF flange parts. ================================================== Notes on the Fischer Core Series Brass Connectors: ----------------------------------------------------- Rev. Date 12-Mar-2020 Current Date 3-Sep-2020 1. The Hermetic version of the Fischer Core Series Brass connectors seems to be the most common type of vacuum feedthrough connector used by this group. 2. So far I have not found a stocking distributor for the Fischer connectors but they do have a US office that I need to contact. A distributor may be Kensington Electronics in Austin Texas. 3. Fischer makes thousands of connector models. For now I will focus on the "104" size in the 16 pin model. 16 pins is about 4 more pins than should be needed. It appears to be standard to use the "A" Polarity, i.e. male contacts in the plug and female contacts in the receptacle. I want solder type contacts (not pcb or crimp). pg 43 The code for 16 pins in a 104 size is: 086 pg 48 In the 104 size the 16 pin model has pin size: 0.7 mm dia pg 48 The 0.7 mm dia pins can take AWG 22 wire (7/30) pg 48 Contact Resistance of 0.7 mm pins 5 mOhm type pg 16 Options, use what appears to be most common: pg 55 Key Code 1, PEEK material, Solder, Chrome: -130 Key Code 1, PEEK material, Solder, Black: -140 Cable Clamp Set for Cable Mound Plug pg 62 Part Number: pg 10 Body Style, Size, Polarity, Contact Config, Options, Cable Mount Clamp 4. For the Cable Mounted Plug the "S" Body Style looks to be the most common. So the part number for this would be something like: S 104 A 086 -130 (or -140) The S 104 connector is 50 mm long overall, 38 mm of its body remains outside of the receptacle, 15 mm outer diameter, max cable OD 8.7 mm, required wrenches of 12 mm and 13 mm. The OD of the 12 mm long section that plugs into the receptacle is ?? mm. 5. The Panel Mounted Receptacle Body Style in a hermetic version is: Rear Front Projecting Projecting ---------- ------------ Front Panel: DEE DBEE Rear Panel: DBPE DBPLE DEE has rear nut and about 21 mm behind the conn lip pg 29 DBEE has rear nut and about 14 mm behind the conn lip pg 30 DBPE has front nut and about 18 mm behind the panel pg 31 DBPLE has front nut and about 7.5 mm behind the panel pg 32 pdf page numbers DBEE will not work as it can only take a 3.5 mm thick panel max. DBPE will not work as it can only take a 4.0 mm thick panel max. The KF flange cove plate is 5.0 mm thick. DBPLE would be best but it could be that only DEE is commonly available. All 4 type have 16x1 threads. A part number may be something like: DBPLE 104 A 086 -130 (or -140) Note that in our application that the DBEE works just as well as the DBPLE because of the thickness of the KF flange cover plate. Note that in our case, if the KF flange pipe stub is longer than 75 mm or so then we can use the DEE case style with no problem and DEE may be the most common. 6. A distributor that does have some of these Fischer connectors is www.cmr-direct.com Basically the only appropriate connectors that they have in stock are a 12 pin set: wire mount plug with pins $151 it is 18 mm x 18 mm x 62 mm and a chassis mount receptacle with sockets $121 it is 27 mm x 27 mm x 32.2 mm with an M18 x 1mm thread. There is also a 24 pin version for $263 for the wire mount plug and $162 for the chassis mount receptacle. 7. Sept 2020 the contact at Fisher is Bryan Presnal and he recommends the distributor: Kensington Electronics in Austin, TX. The contact at Kensington is Daegan Richmond. I have asked Daegan about: S 104 A 086 -130 and DBPLE 104 A 086 -130 but so far I have not heard anything back from him. ================================================== Notes on the Yokogawa GS200 SET Bias Supply: ----------------------------------------------- Rev. Date 21-Feb-2020 1. These supplies do inclue a Guard. It looks like this Guard is not just to block leakage current put also a full Guard to protect against AC Common Mode noise. 2. Note that this Guard can be "ON" or "OFF" where: Guard ON implies that the Guard is connected to the Low Output terminal. Guard OFF implies that the Guard is connected to its own terminal. 3. For instruction book references to the Guard see: page 2-10, 4-5, 4-6. Block diagram is Apx-2. 4. It is still not clear how much capacitive load this supply can take when in Voltage output mode. 5. The SET curve tracer will place almost zero load on these supplies - would they be happier and quieter with a couple of mA load ? ================================================== Cabinet for the Signal Filter Box: ---------------------------------- Rev. Date 17-Mar-2020 1. Each Signal Filter PCB is about 4" x 7" and about 1.75" high so including connectors and such this needs about 6" x 9" by 3" high. 2. The Signal Filter Box has: Input connector e.g. DB-9, Power connector e.g. DB-9, Output connectors e.g. 4x BNCs, Power ON LEDs 3. Possible Boxes: Bud CS-11212-BT 10" x 12" by 6" high also taller version Bud WA-1542 8" x 10" by 8" high also taller version Bud AU-1040 6" x 9" by 5" high LMB Heeger UPS 525-14-7 6.8" x 13.75" x 5" The Bud AU-1040 is Digi-Key No 377-1058-ND for about $30.80 ================================================== Meeting with Niyaz and Johannes Wed. 26-Feb-2020: -------------------------------------------------- Rev. Date 28-Feb-2020 - Correct - we can NOT allow the SET Source to Ground voltage to change when the current through the SET changes so we can not have an RC Filter on the low side of the supply, i.e. we can only AC "clamp" the common side of Voltage and Offset supplies to Ground. - When the system moves from DC operation to RF operation it is not yet clear how much of the DC system will carry over, e.g. do we want read relay to disconnect the DC amplifier components. Also does one need to be able to quickly swap back and forth between DC and RF operation ? - Schedule - we could use the SET DC Curve Tracer starting now. - Money - got an RC number for this work and can start ordering now at up to $1k. - Niyaz is going to measure the length of the KF flange pipe stub on the real cryostat. He reports: Hi Dan The height of the part, to which KF flange will be attached, is 57 mm. Best regards, Niyaz I have been drawing this at 65 mm length. So either a little bit of the Amps will stick down into the open space of the cryostat or else I need to make a short capped stub pipe for the Amps instead of just a flange cover plate. - Everyone will be at the spring meeting next week. - > 1 GHz in pair cable, e.g. SATA - Issue having SET Bias On with Inst Amp Power Off. The SET Bias supplies, e.g. the Offset supply, could pump uncontrolled current into the inputs of the Inst Amps when the Inst Amp power is Off. - Assume that 10e-5 mBar is with pumps running. ================================================== Alternative Hermetic Connector: ------------------------------- Rev. Date 10-Mar-2020 - The Fischer connectors are nice and fancy but they do not appear to be available in normal distribution channels and the ones that are available from the company in England are not actually best suited for the SET DC Curve Tracer application. - An alternative is to use the "standard" micro D-Sub like we have experience with from the L1.5 Cal Trig. This connector type is available from multiple vendors. - On the cryostat use the more expensive hermetic half of the connector from the Cinch Dura-Con series and use the more protected plug with male pins. On the free cable end use the less expensive non-hermetic Cinch Dura-Con series and use the receptacle with female sockets. - In the 15 pin series parts may be: DCDH-15PSBN CONN MICRO-D PLUG 15POS PNL MNT 116-1083-ND $148.50 stock: 20 DCCM15SSB CONN MICRO-D SOCKET 15POS SOLDER 116-1181-ND $34.93 stock: 21 - In the 25 pin series parts may be: DCDH-25PSBN CONN MICRO-D PLUG 25POS PNL MNT 116-1085-ND $168.73 stock: 25 DCCM25SSB CONN MICRO-D SOCKET 25POS SOLDER 116-1185-ND $44.00 stock: 15 - The 21 pin size appears to be less/non standard. - The required flat area for the hermetic half of the connectors is: 15 pin 37.00 mm x 20.00 mm 8.00 mm radius corners 25 pin 45.00 mm x 20.00 mm 8.00 mm radius corners - The actual size of the connectors is: 15 pin 34.16 mm x 17.00 mm 25 pin 41.66 mm x 17.00 mm In both sizes I believe that the ends are semi-circles. ================================================== OPA827 vs OPA828 Final Decision: ------------------------------------ Rev. Date 12-Mar-2020 OPA827 OPA828 ---------- ----------- Date -------- Nov 2006 Sept 2018 Jul 2016 Dec 2018 Power --------- +-4V to +-18V +-4V to +-18V 4.8 mA typ 5.5 mA typ PSRR -------- 25C 0.2 uV typ PSRR 25C 1.4 uV typ PSRR 25C 1.0 uV max PSRR 25C 5.6 uV max PSRR --- 0:+85C 7 uV max PSRR -40:+125C 3.0 uV max PSRR -40:+125C 10 uV max PSRR Bandwidth --------- 22 MHz Gain Bandwidth 45 MHz Gain Bandwidth 28 V/usec 150 V/usec --- RRO Open-Loop --------- 25 C 120 dB min Open-Loop 25 C 120 dB min Open-Loop 25 C 126 dB typ Open-Loop 25 C 130 dB typ Open-Loop --- 0:+85C 117 dB min O-L -40:+125C 114 dB min O-L -40:+125C 114 dB min O-L V OffSet -------- 25 C +- 75 uV typ Vos 25 C +-50 uV typ Vos 25 C +- 150 uV max Vos 25C +-300 uV max Vos --- 0 to +85 C +-350 uV max Vos --- -40 to +125 C +-400 uV max Vos Vos Drift --------- --- 0 to +85 C +-0.30 uv/C typ --- 0 to +85 C +-1.30 uV/C max -40 to +125 C +-0.1 uV/C typ -40 to +125 C +-0.45 uV/C typ -40 to +125 C +-2.0 uV/C max -40 to +125 C +-1.50 uV/C max Bias Current ------------ 25 C +-3 pA typ I Bias 25 C +-1 pA typ I Bias 25 C +-10 pA max I Bias 25 C +-8 pA max I Bias -40 to +85 C +-500 pA max 0 to +85C +-400 pA max -40 to +125 C +-5 nA max -40 to +125 C +-3 nA max Offset Current -------------- 25 C +-3 pA typ Ios 25 C +-1 pA typ Ios 25 C +-10 pA max Ios 25 C +-8 pA max Ios --- 0 to +85C +-500 pA max --- -40 to +125 C +-1.5 nA max Input Noise from Table ---------------------- 0.1 Hz : 10 Hz 250 nVpp typ 0.1 Hz to 10 Hz 340 nVpp typ --- --- 0.1 Hz to 10 Hz 60 nVrms typ --- --- 10 Hz 7.5 nV/sqrt_Hz typ --- --- 100 Hz 4.8 nV/sqrt_Hz typ 1 kHz 4.0 nV/sqrt_Hz typ 1 kHz 4.0 nV/sqrt_Hz typ 10 kHz 3.8 nV/sqrt_Hz typ --- --- 1 kHz 2.2 fA/sqrt_Hz typ 1 kHz 1.2 fA/sqrt_Hz typ Input Noise from Graph ---------------------- 0.1 Hz 59 nV/sqrt_Hz typ 0.1 Hz 62 nV/sqrt_Hz typ 1 Hz 18 nV/sqrt_Hz typ 1 Hz 21 nV/sqrt_Hz typ 10 Hz 7.3 nV/sqrt_Hz typ 10 Hz 7.5 nV/sqrt_Hz typ 100 Hz 4.4 nV/sqrt_Hz typ 100 Hz 4.5 nV/sqrt_Hz typ 1 kHz 3.9 nV/sqrt_Hz typ 1 kHz 4.0 nV/sqrt_Hz typ Actions ----------- Order enough 827 to build 2 Inst Amps (2x 3) plus 2 spare. Order enough 828 to build 2 Inst Amps (2x 3) plus 2 spare. Order engout 145 to build 3 Signal Filters (3x 5) plus 5 spare. ================================================== Connector Pinouts - Power, Cryo, Signal Filter, SET Bias: ----------------------------------------------------------- Rev. Date 6-Aug-2020 This section is a description of and pinout list of all of the cables and connectors in the SET DC Curve Tracer electronics. - Power Source to Load Cables: ---------------------------- These circuits use shielded twisted pair cable to carry the 30 Vpp 25 kHz power from the Power Source box to the various loads in the Inst Amp PS & Bias Filter box and in the Signal Filter box. The cables use DB-9 connectors, male at the source end and female at the load end The Power Source box has female DB-9 connectors. The load boxes have male DC-9 connectors. The Power Cable is an "extension"type cable. Power Cables and Connectors: Pin Function --- ------------------------------ 1 Cable Shield Feed #1 2 One side of the 25 kHz power Feed #1 3 nc 4 One side of the 25 kHz power Feed #2 5 Cable Shield Feed #2 6 Other side of the 25 kHz power Feed #1 7 nc 8 nc 9 Other side of the 25 kHz power Feed #2 Notes: At the Power Source end the Cable Shield is not connected to anything. At the load end the Cable Shield is connected to the local "noisy" ground / chassis ground. 1 2 3 4 5 female 5 4 3 2 1 male 6 7 8 9 solder side 9 8 7 6 solder side - Signal Output Connector on Bias Filter / Inst Amp PS Box Signal Input Connector on Signal Filter Box: -------------------------------------------------------- This path uses DB-15 connectors and carries only the output signals from the Inst Amps to the inputs of the Signal Filters and carries various grounds from the Cryostat to the Signal Filter Box. "V" and "I" Signal Cable IAPS BF Box to Signal Fltr Box: Pin Function --- ------------------------------ 1 spare 2 spare 3 "I" Signal Ground 4 spare 5 spare 6 "V" Signal Ground 7 spare 8 spare 9 spare 10 "I" Signal Cable Shield 11 "I" Signal Cable Center Conductor 12 spare 13 "V" Signal Cable Center Conductor 14 "V" Signal Cable Shield 15 spare The Bias Fltr / Inst Amp PS box has a female DB-15 connector. The Signal Filter Box also has a female DB-15 connector. 1 2 3 4 5 6 7 8 Bias Fltr / Inst Amp PS Box 9 10 11 12 13 14 15 and the Signal Filter Box female Solder Side 8 7 6 5 4 3 2 1 Both ends of Cable 15 14 13 12 11 10 9 male Solder Side - SET Bias Input Connector on Bias Filter / Inst Amp PS Box: ---------------------------------------------------------- This DB-15 connector on the Bias Filter / Inst Amp PS Box receives the SET Bias S-D Voltage and Offset Voltage from the two power supplies that provide the SET Bias. These circuits use is a male DB-15 connector on the Box and a female DB-15 on the long cable that runs over to the SET Bias Power Supplies. SET Bias Supply Cable and Connectors: Pin Function --- ------------------------------ 1 SET Bias Source to Drain Cable Shield & Supply Guard 2 SET Bias Source to Drain Cable High Side 3 nc 4 nc 5 nc 6 nc 7 SET Bias Offset from Gnd Cable High Side 8 SET Bias Offset from Gnd Cable Shield & Supply Guard 9 SET Bias Source to Drain Cable Low Side 10 nc 11 nc 12 nc 13 nc 14 nc 15 SET Bias Offset from Gnd Cable Low Side SET Bias Supply Cable and Connector on the Bias Fltr / Inst Amp PS box has a male DC-15 connector. 8 7 6 5 4 3 2 1 SET Bias Connector on Box 15 14 13 12 11 10 9 male Solder Side 1 2 3 4 5 6 7 8 SET Bias Cable Connector 9 10 11 12 13 14 15 female Solder Side - Cryostat Cable Connector on Bias Filter / Inst Amp PS Box: ---------------------------------------------------------- The SET Bias Filter / Inst Amp PS Box has a DB-25 connector for the short cable that run over to the Cryostat connector. This cable includes about 18 discrete Ground wires that run to the Cryostat Common Ground Point clamp that attaches to the pipe stub on the Cryostat's Inst Amp port. Box to Cryostat Cable and Connectors: Pin Function --- ------------------------------ 1 Cryo Ground to Inst Amp Power Return 2 Inst Amp Power +10 Volt 3 Cryo Ground to "V" Signal Filter 4 "V" Signal Cable Center Conductor 5 "V" Signal Cable Shield 6 Cryo Ground to SET Bias Filter 10 uFd cap High side 7 SET Bias HIGH into Cryostat 8 Cryo Ground to SET Bias Filter 100 uFd cap High side 9 Cryo Ground to SET Bias Filter Zeners and Bleaders 10 "V" Amp Thermistor 11 spare 12 spare 13 Cryo Ground to Shield on SET Bias S-D Cable 14 Inst Amp Power -10 Volt 15 Cryo Ground to Inst Amp Power Return 16 Cryo Ground to "I" Signal Filter 17 "I" Signal Cable Center Conductor 18 "I" Signal Cable Shield 19 Cryo Ground to SET Bias Filter 10 uFd cap Low side 20 SET Bias LOW into Cryostat 21 Cryo Ground to SET Bias Filter 100 uFd cap Low side 22 Cryo Ground to SET Bias Offset Low side 23 "I" Amp Thermistor 24 Cryo Ground to Meter Selector Switch 25 Cryo Ground to Shield on SET Bias Offset Cable Female connector on the Bias Fltr / Inst Amp PS Box Male connector on the Cable to the Cryostat Cable to Cryostat male Solder Side 13 12 11 10 9 8 7 6 5 4 3 2 1 25 24 23 22 21 20 19 18 17 16 15 14 Bias Fltr / Inst Amp PS Box female Solder Side 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 - Cryostat Hermetic Connector: ---------------------------- The connector to the Inst Amps inside the cryostat is mounted on the cover plate for a KF Flange on a 40 mm ID pipe stub. Cryostat Hermetic Connector Pinout: Pin Function --- ------------------------------ 1 Cryo Gnd to Inst Amp Power Return 2 Inst Amp Power +10 Volt 3 Cryo Gnd to "V" Amp Ground 4 "V" Amp Output 5 Cryo Gnd to "V" Amp Reference Ground 6 SET Bias HIGH 7 spare 8 "V" Amp Thermistor 9 Inst Amp Power -10 Volt 10 Cryo Gnd to "I" Amp Ground 11 "I" Amp Output 12 Cryo Gnd to "I" Amp Reference Ground 13 SET Bias LOW 14 spare 15 "I" Amp Thermistor At this end of the Cryo Cable there are additional Cryostat Ground connections to the: - Power Cable Shield - SET Bias Cable Shield - "V" Signal Cable Shield - "I" Signal Cable Shield The connector on the KF Flange cover plate is Hermetic. It is a 15 pin plug with male pins. 8 7 6 5 4 3 2 1 Cryo Connector 15 14 13 12 11 10 9 Solder Side The matching connector is on the free end of a short cable that runs over to to box that contains the Inst Amp Power Supply and the SET Bias Filter. This connector is a receptacle with female sockets. 1 2 3 4 5 6 7 8 Cable Connector 9 10 11 12 13 14 15 Solder Side ================================================== NTC Thermistor Resistance vs Temperature: ----------------------------------------- Rev. Date 13-Apr-2020 The following table lists the expected relationship between the temperature in degrees Centigrade of the NTC Thermistor on the Instrumentation Amplifier and its resistance in Ohms. The initial calibration of the Thermistor should be within 1% of the expected resistance at 25 deg C. Temperature Resistance Temperature Resistance Degrees C Ohms Degrees C Ohms ----------- ---------- ----------- ---------- -55 446.05 k 0 27.33 k -50 332.81 k 5 22.10 k -45 250.44 k 10 17.97 k -40 190.03 k 15 14.70 k -35 145.36 k 20 12.09 k -30 112.06 k 25 10.00 k -25 87.04 k 30 8.31 k -20 68.10 k 35 6.94 k -15 53.67 k 40 5.83 k -10 42.58 k 45 4.91 k -5 34.00 k 50 4.16 k ================================================== The Following Are Estimates of the Bias Filter Loss at DC: ---------------------------------------------------------- Rev. Date 13-Apr-2020 Leakage in the large film capacitors will cause a small DC voltage drop across the series resistors in the SET Bias Filter. The following are estimates of the maximum and typical loss at DC in the SET Bias Filter. Polypropylene Film Capacitor Leakage: type R75 10 uFd > 30k Meg Ohm x uFd min > 150k Meg Ohm x uFd typ ---> 3k Meg Ohm min 15k Meg Ohm typ Polyester Film Capacitor Leakage: type R60 100 uFd > 1.25k Meg Ohm x uFd min > 5k Meg Ohm x uFd typ ---> 12.5 Meg Ohm min 50 Meg Ohm typ So in the first stage of the Bias filter with a 1k Ohm series resistor and a 100 uFd polyester capacitor the DC leakage is expected to be: 80 ppm maximum In the second stage of the Bias filter with a 25k Ohm series resistor and a 10 uFd polypropylene capacitor the DC leakage is expected to be: 8.3 ppm maximum ================================================== Meter Switch on the SET Bias Filter - Inst Amp P.S. Box: ---------------------------------------------------- Rev. Date 16-Apr-2020 The intent is to have a pair of tip jacks and a Meter Selector rotor switch on the SET Bias Filter - Inst Amp Power Supply Box so that a hand-held multi-meter can be used to check many of the basic functions of this box to insure that the overall SET DC Curve Tracer is working correctly. The Meter Selector Rotor Switch has the following 8 positions: 1. Inst Amp VCC Supply Voltage expect about +10 Volts 2. Inst Amp VEE Supply Voltage expect about -10 Volts 3. V Inst Amp Thermistor Resistance expect about 4k : 450k Ohms 4. I Inst Amp Thermistor Resistance expect about 4k : 450k Ohms 5. V Inst Amp Output Voltage expect about -1V : +1V 6. I Inst Amp Output Voltage expect about -1V : +1V 7. SET Bias Filter Offset from Ground Input Voltage about -3V : +3V 8. SET Bias Filter Source to Drain Input Voltage about 0V : +10V This will require an 8 Position (or more for spare or Off positions) two pole rotor switch. All sources will be isolated from the switch with resistors so that a fault in the hand-held multi-meter will not cause damage to the SET DC Curve Tracer electronics. 1k Ohm isolation resistors for the power supplies test points and 10k Ohm isolation resistors from the Instrument Amplifier outputs. ================================================== Top Drawer Plotting of SET Electronics Characteristics: ------------------------------------------------------- Rev. Date 21-Apr-2020 I installed Top Drawer on the Linux Centos 7 machines. It appears to work almost the same as the "good old" VMS and Unix Top Drawer. The Linux version came from Fermilab a as 64 bit statically linked executable so it should be all self contained. The source URL including the documentation is: https://home.fnal.gov/~parke/TD/ It appears that I need to give the take command file the extension ".top" for this version to work correctly. The ps2pdf is already on the Centos 7 machines and appears to work as normal. Example run: [dan@hubmac TD_Plot_Work]$ [dan@hubmac TD_Plot_Work]$ ../TD/td64 set_100_hz_one.top Plot 1 Done. [dan@hubmac TD_Plot_Work]$ # and I have set_100_hz_one.ps [dan@hubmac TD_Plot_Work]$ [dan@hubmac TD_Plot_Work]$ ps2pdf set_100_hz_one.ps [dan@hubmac TD_Plot_Work]$ # and I have set_100_hz_one.pdf ================================================== ==================================================