Home
The Test, Usage and Maintenance of Power Switching
Contents
1. Resistance O 000e O00 1 068e 000 3 000e 000 Resistance 0 000e 000 6 357e 001 3 000e 000 Resistance O 000e 000 1 87ze 000 32 000e 000 Resistance O 000e 000 1 649e 000 3 000e 000 Resistance O 000e o000 5 745e O01 3 000e 000 Resistance 0 000e 000 2 774e 000 3 000e 000 Resistance O 000e 000 1 009e 000 3 000e 000 Resistance O 000e O000 1 640e 000 3 000e 000 Resistance O 000e tO00 5 Z0 e OOl1 3 000e 000 Resistance O 000e 000 1 107e 000 3 000e 000 Resistance 0 000e 000 7 926e O01 3 000e 000 Resistance O 000e to00 1 316e 000 3 000e 000 Resistance O 000e 000 1 405e 000 3 000e 000 Resistance 0 000e 000 6 235e O01 3 000e 000 Resistance O 000e 000 7 303e O01 3 000e 000 Resistance O 000e 000 7 505e O01 3 000e 000 Resistance 0 000e 000 7 726e O0O1 3 000e 000 Resistance O 000e O00 1 35e 000 32 000e 000 Table 7 Relay measurement results before cleaning routine Resistance 0 000e o000 4 420e 001 3 000e 000 Resistance 0 000e 000 4 551le 001 3 000e 000 Resistance O 000e too00 4 7 60e O01 3 000e 000 Besistance 0 000e 000 4 915e 001 3 000e 000 Resistance 0 000e 000 4 6 762 001 3 000e 000 Resistance 0 000e 000 S 168e 001 3 000e 000 Resistance 0 000e 000 5 062e 001 3 000e 000 Resistance 0 000e 000 4 9246 001 3 000e 000 Resistance 0 000e 000 4 59 76e 001 3 000e 000 Resistance 0 000e 000 4 55le 001 3 000e 000 Resistance 0 000e 000 4 656e 001 3 000e 000 Resistance O 000e oo00 4 5646 001 3 000e 000 Besistance 0
2. e Test Limit 12v 100 25 Q 1 Q 619 mAmps D Relay Cleaning In the power relay specifications section the effect that fritting has on clearing the relays of contamination was discussed In some situations the buildup of contaminants on power relays can become significant enough that a method is required to rectify the problem To deal with this issue relay vendors recommend periodic cleaning through the use of arcing From a Tyco application note Controlled arcing of short duration can be beneficial in achieving the rated life of the contacts Unlike fritting in which minimum levels of voltage and current are applied to a closed relay arcing is achieved through hot switching of the relays with higher levels of voltage and current applied The appropriate voltage and current levels are based on the type of material used in the relay contacts The first step in designing an approach to generate a cleaning arc is to determine the metals that make up the relay contacts so that the specific levels of voltage and current necessary to develop an arc can be defined It may require contacting the switch vendor for a relay datasheet if the information is not available in the switch module user s manual It may also be possible to track down the appropriate relay data sheet by getting the relay part number off of the relay as shown in Figure 3 an actual switch module relay photo Contact Data Arrangements 1 Form
3. 000e 000 S 018e 001 3 000e 000 Resistance 0 000e 000 5 324e 001 3 000e 000 Resistance 0 000e 000 4 92 1le 001 3 000e 000 Resistance 0 000e 000 4 75le O0O1 3 000e 000 Resistance 0 000e 000 4 54246 0001 3 000e 000 Resistance 0 000e 000 4 590e 001 3 000e 000 Resistance 0 000e 000 4 56le 001 3 000e 000 Resistance 0 000e 000 4 674e 001 3 000e 000 Table 8 Relay measurement results after cleaning routine H Symptoms Indicating Relay Cleaning is Necessary When oxidation builds up on switch contacts various symptoms can indicate that a cleaning cycle is warranted 1 Failures in TPSs start appearing on previously working TPSs or there are self test errors reporting path continuity problems These may be hard errors or they may be intermittent failures since each closure of the affected switches can make varying level of contact 2 If you connect a meter across the various switches then you may observe a Resistances over the rated path resistance usually more than an ohm b If the relay is cycled opened and closed different resistance values are seen and can be very inconsistent with variations of 100s of mOhms to ohms On a clean relay you will usually see reading within 10 or 20 mOhms on a switch cycle c If you leave the meter attached you may see the readout drift lower over time although the drift can reverse and go back up Regardless the drift can be over the range of many ohms I Effects of Hot Switchi
4. 1x10 Q MINIMUM CONTACT RATING 10 mV dc Table 1 Vendor signal switch specification sheet The next level of switches are low power switches which operate from 3 amps to 5 amps It is important to note that although switches with these specifications are sometimes used for signal transmission doing so uses the switch outside of its specified operating area This type of switch will have a minimum contact rating that is higher than the signal switch a rating of 5V and 10 mAmps is common These switches with maximum current values between 3 amps and 5 amps may or may not have gold plated contacts Relays that operate in this range may offer gold plating as an option but this option is transparent to the end user and 1s dictated by the switch module vendor which adds some difficulty for the test engineer when determining if a cleaning approach should be employed One characteristic of this type of switch is that if the contacts are not gold plated their ability to deal with signals below the minimum contact rating 1s not very good But when the switch contacts are gold plated the switch transmission of these below minimum contact ratings signals can be quite successful The reason for this is that the gold plated contacts if not abused with hot switching will be more resilient to oxidation build ups and the quality of the transmission of low power signals will be more consistent over time If the contacts are not gold pla
5. displaying intermittent failures which are difficult to diagnose When testing or troubleshooting power switch paths the engineer has to consider the limitations of using standard DMMs to make path resistance measurements Using a DMM will provide some basic information but only readings that indicate a total closed path shorted or an open can be trusted Readings that vary from 1 ohm to 100 ohms could be caused by oxidation on the relays not overcome by the current and voltage from the meter The best approach for testing power switch paths is to use DC power supplies which provide the minimum levels of voltage and current required to ensure a solid relay closure To get the longest possible life from power switches and to obtain consistent test results over a long period of time a relay cleaning program along with the necessary fixturing and software are needed Cleaning should be done at intervals which are dictated by the test system s usage and the cleaning approach should be adjusted based on feedback from system users regarding any TPS or self test intermittencies V REFERENCES 1 Wikipedia Definition of Relay http en wikipedia org wiki Wetting current 2 VTI Corporation SMP3001 SMP5005 and SMP2001 Specification Sheet SMP2 Series User s Manual July 2004 3 Tyco Electronics Corporation Relay Contact Life http relays tycoelectronics com appnotes app_pdfs 13c3236 pdf Wetting Current 4 Tyco Electronics C
6. voltage and current read back capabilities required by this self test approach In the test configuration illustrated in Figure 2 a DC power supply with internal measurement capabilities 1s connected in a configuration which allows testing of 5 individual switch modules A load resistor is placed across the DC power supply allowing the power supply to be tested before any relay verification is done In the design a precision 0 5 100 ohm 25 watt load is chosen to allow the best combination of voltage and current verification of the supply The supply used is a 35V 8 5 amp supply so the full voltage of the supply 35V can be verified and the current can be tested to approximately 350 mAmps If the power supply test fails then no switch tests are run An important aspect of having the 100 ohm load across the DC power supply 1s that it provides the means to factor out the error in the power supply measurement circuitry This test approach uses the DC power supply to measure the current passing through the relay test paths but on many power supplies the amount of specified error can be significant enough so that using a DC power supply is impractical in some cases reaching 80 error To overcome this problem an initial measurement is made by setting the DC power supply to the test voltage that will be used in the relay test which in this case is 12 volts In an ideal set up this would result in 120 mAmps across the 100 ohm load By using the
7. A SPSENO Wiring Diagram Code 1 2 3 pe 2 Form A DPSENO Wiring Diagram Code 5 1 Form C SPDT Wiring Diagram Code 1 2 3 2 Form C DPDT Wiring Diagram Code 5 Material Silvernickel 90 10 Minimum Load 12V 100mA4 Expected Mechanical Life 10 million operations Initial Contact Resistance 100 miltohms max 1A 12VDC ee NO 5VDC 5A 250VAC _ 5A 20v0C 09881 Figure 3 Spec sheet showing relay contact material Once the contact material is known Table 5 can be referenced to determine the arcing voltage and current necessary for the relay to be cleaned For example the contact material of the relay in Figure 3 consists of a combination of silver and nickel Based on the material with the higher values in this case nickel by selecting a voltage of 14V and a current of 0 5 amps an arc will be generated Conductivity Voltage Voltage Current Table 5 Relay arcing voltage and current values From Tyco app Relay Contact Life 3 Palladium Tungsten Knowing the arc voltage and current a load must be chosen which will produce the required arc current when the power supply is programmed to output the arc voltage In this case using a 28 ohm load would produce the exact current of 0 5 amps desired when the power supply is set to 14V Since these values do not have to be exact opting for a more common 25 ohm 25 watt load resistor is a good option This will provide a current of 560 mAmps from the power
8. The Test Usage and Maintenance of Power Switching Subsystems Kevin Paton Teradyne Inc North Reading MA Kevin Paton teradyne com Abstract Unlike signal switching power switches have significant minimum operational requirements for both voltage and current These minimum requirements affect usage in test programs and self test approaches This paper will discuss the considerations necessary when testing power switches or when using power switching in test programs In addition techniques will be described which can reduce or remove build up of deposits improving switch operation I INTRODUCTION Unlike signal switching power switches have significant minimum operational requirements for both voltage and current minimum requirements affect usage in test programs TPSs and self test approaches In addition to TPS usage and testing considerations power switches must also be maintained since they have a tendency to build up deposits on their contacts which can degrade the characteristics of the switch However there are techniques which can reduce or remove build up of deposits improving switch operation Il POWER SWITCHING SPECIFICATIONS When creating a system self test or laying out the paths to be used in a test program the test engineer needs to consider the types of signals or power that will pass through the various switches to determine if the switch specifications line up with the actual requirements of the self te
9. arts looking at switch module self tests within a system they produce a layout in which they bring a DMM to the receiver interface and use a jumper pattern across the switches to measure the switch module relays for both opened and closed resistances This approach works well for signal switching but when testing power switches using a standard DMM approach can lead to problems As shown in Table 2 the minimum current needed to ensure relay closure on that particular power switch was 100 mAmps and 12V However when making resistance measurements the typical DMM uses voltages and currents that are much smaller than the switch s minimum contact rating Table 3 shows the current and voltage values used by several popular COTs instruments when measuring resistance at the lowest meter range the range used to make the closed relay measurements Note that none of these modules come close to the 12 volts 100 mAmps necessary to meet the minimum contact resistance specification for this switch Meter Voltage Teradyne Ai760 100 mV Agilent 1412 100 mV Agilent 3458 Teradyne Ai710 100mA 500 mV Table 3 Current and Voltage Levels Used by Various DMMs When Making Low Range Resistance Measurements Given this information the only practical way to reliably test power switches is to use a DC power supply to source the required voltage and current levels Most testers offer some form of off the shelf DC power supply that provides internal
10. at affects these ratings is the relay contact material Switches that handle up to 3 Amps will typically have some form of gold plating on their contacts If is the switch supports between 3 Amps and 5 Amps the gold plating tends to be an option but above 5 Amps the switch contacts will be constructed of some other more robust metal such as silver or tungsten which is not quite as good a conductor as gold but can tolerate the higher currents being switched The following three tables contain specifications for different switch types which illustrate how the minimum contact rating can vary based on the current carrying capabilities of the relays involved These tables contain specifications for signal switching with very low minimum contact ratings to high power switching with much higher current and voltage minimums Table 1 shows the specifications for a signal switch module Note that the maximum current value is only 2 amps and the minimum contact rating of the signal switch is very small at 1OmV and 10uA GENERAL SPECIFICATIONS MODEL TYPE Multiplexer CHANNELS 64x 1 VXI DEVICE TYPE Register Based VXIPLUGE PLAY WIN9S NT Yes SWITCHING TIME lt 3 ms RATED SWITCH OPERATIONS p Mechanical 1x10 Electrical 1x 10 Full Load POWER SPECIFICATIONS MAXIMUM SWITCHING V OLTAGE 300 V ac rms 300 V de MAXIMUM SWITCHING CURRENT 2A MAXIMUM SWITCHING POWER 60 W dc 125 VA DC PERFORMANCE PATH RESISTANCE 0 59 INSULATION RESISTANCE gt
11. avoided Note The possibility of using 5A power switches as signal paths is not recommended but is raised here because it is not uncommon to see this in real world applications This approach has potential issues which are discussed in the section entitled Real Life Examples Problems and Considerations G Examples of the Affects of Relay Cleaning The results in Table 7 and Table 8 show measurements made on a 5A power switch before and after a cleaning cycle Most of the pre cleaning measurements are well over 500 mOhms with one at almost 3 ohms After cleaning all of the readings are around 500 mOhms or better Through experimentation it was found that after a couple hundred cycles the measurements were back to the pre cleaning state What is interesting about this demonstration is that by cleaning the user could make measurements well under the minimum contact rating value for a while The measurements in Table 7 and Table 8 were made with a standard DMM so no fritting takes place during these measurements Although the DMM does not meet the minimum contact rating for this demonstration it provides a more tangible affect of the cleaning If a measurement approach was used that did meet the minimum contact rating it could be assumed that the results in Table 7 would be significantly better but would mask the cleaning affect Resistance O 000e tO00 1L 19 e 000 3 000e 000 Resistance 0 000e 000 amp 31 e OoOl1 3 000e 000
12. itially to become intermittent e The power switch possibly has gold plating on the contacts which if used in other TPSs to pass higher currents and or were being hot switched could eventually burn off causing the path transmission characteristics to significantly degrade over time to the point that the paths can no longer support TTL signal transmission Logic State Output Input 2 0 V Current High 400 uA Current Low 16 mA 1 6 mA Table 9 Power and current levels for TTL signals In this situation the engineer can either rewire the fixture and rewrite the TPS to use signal switching or use a regular relay cleaning regimen The cleaning techniques were described in the previous section with the end result being that the contaminants that can build up on relay s contacts can be removed allowing the switch to work at levels below the minimum contact rating This customer opted to rewire their fixture which was the recommended approach IV CONCLUSION When laying out test programs or self tests the test engineer needs to take into account all switch module specifications including the often ignored minimum contact rating specification to ensure that the best available paths are used for the signal and power routing Using power paths for signal transmissions should absolutely be avoided A TPS using this approach may appear to work during the initial development cycle it could very well fail weeks or months down the road
13. ng Although the cleaning process involves hot switching the effects on the switch life should not be adverse and in fact should help to extend the switch life The cleaning approach described above involves currents which are relatively low around 2 amps The graph in Figure 10 demonstrates the effect of hot switching on a common relay family that is used on 5 Amp switch modules readily available for test systems It is clear that hot switching at certain currents can significantly reduce the life of the particular switches while at the levels being used to accomplish the cleaning process 0 5 amps typical degradation is not a factor Contact curent amp Figure 5 Effect of hot switching on relay life III REAL LIFE EXAMPLES PROBLEMS AND CONSIDERATIONS A TPS Failure Leads to More DMM Induced Failure Symptoms A customer reported that his TPS would not run consistently He stated that he could connect up his handheld meter across the switching being used by the TPS and see elevated readings that were inconsistent and tended to float around therefore he suspected the connections to his power relays were the cause He re terminated several of the receiver pins involved and this did not changed the results A field service engineer was dispatched to investigate and used the system self test to diagnose the problem It was found that there were two faulty pins at the switch module interface and that the pins the TPS de
14. orporation Power Switching Applications for High Voltage Relays Graph showing hot switching effects http relays tycoelectronics com kilovac power 5 Porter Brumfield Spec sheet link RT Series Potter Brumfield Relay Specification http relays tycoelectronics com pnb asp click on RT DC Coil 8 16A One or Two Poles 6 Omron Corp 5A Slim socket mountable PCB Relay http www components omron com components web webfiles nsf FILE S family html ID TABE 72ZR5S 7 Tyco T90 relay spec sheet http relays tycoelectronics com pnb asp click on T90 30A 1 Pole DC Coil PCB Relay 8 Agilent 3458A DMM data sheet http cp literature agilent com litweb pdt 5965 4971E pdf 9 Agilent 1412 DMM data sheet http cp literature agilent com litweb pdt 5965 5563E pdf
15. s ganged together to accomplish a cleaning circuit Note that in the previous Self Test section all the self tests are run at lower voltages while for the relay cleaning the voltages are increased This keeps the arcing limited to the cleaning procedure only Calculations for Cleaning Currents and Voltages Vin PS Rsw Rlid I I I Power Power Load Rsw Rld Rsw Total Rsw T 14V 100Q 25Q 20Q 140 560 700 7 8 W 9 8 W mA mA mA 15V 100 Q 30Q 23Q 150 500 650 7 5W 9 75 mA mA mA W 18V 100 Q 35Q 26Q 180 514 694 93W 12 5 mA mA mA W mA mA mA W mA mA mA W W Table 6 Calculations for the first two modules being tested using layout in Figure 2 F Gold Plated Relays When considering cleaning power switches it is important to know that if the switch handles 5A or less there is a chance that the relay contacts are gold plated A gold plated contact resists oxidation and is a superior conductor Because of this switches using gold plated relays tend to work well below the minimum contact rating allowing even TTL signals to pass If the contacts are gold using the arcing process could cause the gold plating to burn off of the contacts resulting in degraded operation at levels below the relay s minimum contact rating If the switching will be used at levels over the minimum contact rating this is not an issue but if the switches will be used to pass lower level signals hot switching should be
16. same test voltage and highly accurate load resistors the resulting current measurement across the 100 ohm load will provide an offset to be used for all relay path measurements For example if the calculated expected value is 120 mAmps but the initial test result is 100 mAmps adding 20 mAmps to all relay test result calculations will factor out the power supply and load error M SMP2001A gt MATRIX MODULE 1 MATRIX MATRIX a a a a MODULE 2 MODULE 4 MATRIX MATRIX MODULE 5 MODULE 3 K7 KS oe KI ee K10 o lt 25 Ohms S 30 Ohms S 25watts gt 35 Ohms lt 25 Watts lt 25 Watts gt 45 Ohms 2 gt 500hms lt 25 Watts lt 25 Watts 100 Ohms 25 Watts Ne T Figure 2 Self Test Layout using Single DC PS to Test 5 Modules The test configuration in Figure 2 allows the 20 switches per module to be tested by one power supply This configuration is robust however if a path is stuck closed an alternate means must be used to determine which switch module is faulty To help isolate this type of failure down to the module level different load resistors are used for each module under test If there is a stuck closed path the module will be indicted based on the calculated path resistance This would be caught during the initial test when the power supply is tested standalone across the 100 ohm load If a switch path is shorted the load associated with that module
17. signer had flagged with his manual measurements were actually in good working order The handheld meter the designer had used delivered 1 mAmp at 100 mV to measure a power switch that had a minimum contact rating of 500 mAmps at 12 V To effectively trouble shoot these switch paths the customer would have needed a power source with a load in series that would create 1 2 amp of current required by the switch he was measuring With this set up the hand held meter could be used in the DC voltage mode to take a voltage measurement that could be used to calculate the resistance B Intermittent TPS Results It is not uncommon to find power switches particularly lower level 5A power switches used as paths for various signals and for connections to make resistance measurements in addition to their intended purpose of passing power This practice is highly discouraged Related to this situation a customer reported that their test program which had operated effectively for months was now displaying intermittent results It was determined that the test used power switches to transmit TTL signals Table 9 shows the voltage and current levels for TTL signals and one can see that many of the logic levels do not have the voltage and current levels needed for fritting to take place Some factors could cause this type of failure e Deposits caused by oxidation increase the path impedances over time potentially causing a program that worked reliably in
18. st or the test program When reviewing switch specifications the usual ratings that one would investigate include capabilities such as maximum power maximum voltage maximum current and path resistance However one specification that often goes unnoticed is the minimum contact rating From Wikipidea The minimum contact rating is based on what is known as the wetting current or sealing current is the minimum current needing to flow through a newly closed mechanical switch or relay in order to break through any film contact oxidation that may have been deposited on the switch contacts while open The film contact oxidation occurs often in areas with high humidity 1 The reason this rating is important is that any time a signal either complex or DC is transmitted through a switch and that signal has current and or voltage levels less than the switch s minimum contact ratings it can lead to unreliable operation the reasons for this will be described below Yet many engineers do not even consider this rating Some switch vendors even leave this rating off of their published switch specifications 978 1 4244 9363 0 11 26 00 2011 IEEE however it is critical that the test or TPS designer obtain this specification because it is an important factor in the design of the test If it is not in the standard published datasheet then the switch vendor should be approached to supply this specification One key characteristic th
19. supply and is sufficient for this cleaning approach The cleaning approach itself can be an automated program with a formal fixture or if the need for cleaning is infrequent a manual set up can be built using the switch module s soft front panel to manually control the switches The photo in Figure 4 is an actual set up used to manually clean switch modules In this case power is brought from an onboard DC power supply to the switch interface via jumpers that contain the required load A daisy chain of jumpers is placed across the switches with the high lead on one side of the relay string and the low lead on the other Since there is no requirement to test the relays it is not necessary to isolate the individual relays which makes a daisy chain approach the preferred method Figure 4 Relay cleaning setup E Automated Example Figure 2 shows the schematic used for the self test but now the setup will be used to clean the five individual switch modules Table 6 shows the load resistors used and the power and current calculations Note that the voltage is kept at a minimum of 14V and the current at 0 5 amps or greater to meet the arcing level requirements derived from Table 6 This design example would be overly complicated if it was not a dual purpose approach If this was to be used only for relay cleaning the design could be simplified All load resistors with the exception of the 25 ohm load could be eliminated and the switche
20. ted then the transmission capabilities can deteriorate over time although this situation can be remedied if a cleaning technique is employed Table 2 contains the specifications for a high current power switch Note that the maximum current is 16 amps and the minimum contact ratings are much higher at 12V and 100 mAmps It is not uncommon for power switches to have a minimum contact current rating as high as 1Amp GENERAL SPECIFICATIONS MODEL TYPE Power CHANNELS 20 SPST VXI DEVICE TYPE Register Based VXIPLUG amp PLAY WIN95 NT Yes SWITCHING TIME lt 10 ms RATED SWITCH OPERATIONS x Mechanical 1x 10 Electrical 1 x 10 Full Load POWER SPECIFICATIONS MAXIMUM SWITCHING VOLTAGE 400 V ac 125 V de MAXIMUM SWITCHING CURRENT 16 A MAXIMUM SWITCHING POWER 300 W dc 4000 VA DC PERFORMANCE PATH RESISTANCE lt 0 1Q INSULATION RESISTANCE gt 1x10 Q2 MAXIMUM THERMAL OFFSET CHANNEL HI LO lt 50 uV ISOLATION gt 1x10 2 N 12 V dc 0 1 A Table 2 Vendor signal switch specification for a high current power switch A Minimum Contact Rating The materials used to construct power relay contacts are typically metals such as tungsten or palladium and alloys such as silver nickel These metals tend to oxidize in a very short period of time even in a sealed relay This oxidation increases the resistance across the contacts so it is important that it is removed for optimal operation of the switch To do this a minimum voltage and c
21. urrent combination must be applied which causes electrical and thermal destruction thus eliminating the layer of contaminants This electrical destruction is in the form of an arc or a spark at lower voltage and is known as A fritting The electrical destruction forms a thin hole in the oxidation layer which allows current to flow The resulting current flow is known as B fritting and this flow generates the thermal destruction that removes the rest of the contaminants Figure 1 shows a graphical depiction of the electrical and thermal destructive effect In this graph the two types of fritting are clearly demonstrated along with the affect on the impedance across the switch contacts On this particular switch A fritting occurs when the signal across the contacts hits approximately 330 mV and completes when the current hits approximately 0 4 amps at which point the voltage level drops to a steady 0 18 V regardless of the level of current that is supplied This graph also depicts the affect that the fritting has on the path resistance Initially there are 1 5 ohms across the switch but after A fritting this immediately drops to around 0 5 ohms As more current flows B fritting causes the contact resistance to drop even further Note the initial resistance of 1 5 ohms drops to 19 mOhms after fritting From Tyco Automotive app note _ A fritting Figure 1 Affect of fritting B Self Test Typically when a designer st
22. would then be in parallel with the power supply s 100 ohm test load and the resulting current allows the test program to mathematically determine the faulty module Because of the possibility of a shorted path it is important that a reasonable current limit value is chosen during this initial test to eliminate the possibility of damage to the test circuitry or the module under test C Test Limits When calculating the test limits the first step is to determine the expected current through a perfect path Table 4 contains values particular to the first module in Figure 2 which has a 25 ohm load attached resulting in an expected current of 0 60 amps 12V 100Q 25Q If the maximum relay path resistance to be allowed is 3 ohms the low limit would be e Test Limit Vin R_PS_ load R module load 1 R _load error R_test_ limit e Test Limit 12v 20 Q 1 01 2 Q e Test Limit 541 mAmps Calculations for Relay Testing PS Curr Curr Curr Power Power 25 120 480 600 Table 4 Key Voltages and Currents in the Test Circuit On the high limit side the only factor which should allow the measured current to go above the 0 60 amp expected current would be an error in the 25 ohm module test load and since all other sources of error would have been factored out in the initial current test This would occur if the 25 ohms load was less than 25 ohms Using a 1 Q limit would produce an upper value of 619 mAmps
Download Pdf Manuals
Related Search
Related Contents
iZotope Ozone 5 Help Documentation Electronic Arts 14633194579 Video Games User Manual PDF Saab 900 - Saab Sport Club D-GRADE エアーパージセット 形式:FGS,FGT Mode d´emploi SCHICK - Fraiseuse S1 Dell OpenManage Network Manager Version 5.0 Web Client Guide Copyright © All rights reserved.
Failed to retrieve file