Home
Help
Search
Login
Register1
Repair / Re: trying to identifying parts
« Last post by fzabkar on Today at 05:42:03 pm »"U56" = 0.56uF = 560nF ceramic capacitor ??
Recent Posts
2
Test Equipment / Inlet power filter grounding considerations
« Last post by Nevada on Today at 05:40:49 pm »Hi,
I have some older test equipment with "classical" power input filter (by Schaffner), without ground pin. The device is double insulated, so it can't be grounded to the PE (AFAIK).
How can I replace the filter correctly? If I simply buy a new filter with the usual IEC socket (L/N/PE), then I technically ground the device to the PE, since it has metal case, and the filter has also a metal shield, which is directly connected to the ground (PE pin). The other problem is that the circuit in the new filters is also connected to the metal shiled, which means that at least one or two capacitor is present between the L and the metal housing. I feel that I can't simply put a new (grounded) filter in it if I don't want to break the double insulation rules.
I have some older test equipment with "classical" power input filter (by Schaffner), without ground pin. The device is double insulated, so it can't be grounded to the PE (AFAIK).
How can I replace the filter correctly? If I simply buy a new filter with the usual IEC socket (L/N/PE), then I technically ground the device to the PE, since it has metal case, and the filter has also a metal shield, which is directly connected to the ground (PE pin). The other problem is that the circuit in the new filters is also connected to the metal shiled, which means that at least one or two capacitor is present between the L and the metal housing. I feel that I can't simply put a new (grounded) filter in it if I don't want to break the double insulation rules.
3
Projects, Designs, and Technical Stuff / Re: Digital weigh balance - DIY
« Last post by S. Petrukhin on Today at 05:39:16 pm »When we install sensors for legs at industrial facilities, we sell kettlebells in a set so that the staff periodically calibrates.
Or install reset line to zero on the scales according to the void sensor if it possible.
I do not know the exact data, not involved in this process, but according to colleagues, there is information that the readings of the scales lose about 1% per winter storage of grain - they do not return to zerro after released.
This makes sense. So paulca would have to periodically reset the zero or recalibrate, the same as would be necessary with the elastic bumper idea.
All depends from required accuracy. I suggested a couple of solutions there that do not require calibration.
4
Test Equipment / Re: New Hantek DSO2X1X models?
« Last post by Poogo on Today at 05:34:28 pm »Hi everyone, long time no see!
https://www.eediscuss.com/forum.php?mod=viewthread&tid=19361&extra=page%3D1
https://www.eediscuss.com/forum.php?mod=viewthread&tid=19361&extra=page%3D1
5
Mechanical & Automation Engineering / Re: High density connector troubleshooting
« Last post by DiodeDipShit on Today at 05:31:49 pm »Correction to above: Liquid with a brush in the lid AND markers and they have a spray removal solution.
6
Mechanical & Automation Engineering / Re: High density connector troubleshooting
« Last post by DiodeDipShit on Today at 05:29:24 pm »Anything is a tool for Mechanical engineering. We have used wood for quick prototypes, stickgum and cigarette wrappers for insulators and shims, Silly Putty for shaping and molding, I bought some of my best tiny tools at Sewing Machine stores, Art supply stores and Fabric stores. You can even get a fair, inexpensive little CNC machine for quick and accurate R&D of insulators, shims or gaskets from Cricut. Look at the Marker 3.
Carbon paper and Dykem are common to analyze and refine interfering fit. Dykem comes in spray cans, Liquid with a brush in the lid markers and they have a spray removal solution. It sprays very thin if used properly.
This may be Your best bet to find out if the connectors are being poorly aligned, have too much pressure at specific points between the male and female connectors meaning the issue may be in manufacturing and not Your fault.
I suggested alignment pins for the opposite side of the connector and receiving holes would have to be drilled into the mating connector. These would have to be done by a very reliable machine shop.
In regards to; the existing pins may be to small allowing useless 'precision' due to the mating part rocking. Try a couple of descending thickness split shims to connect the terminals in stages, this will reduce some of the rocking. You may also try using an fixtured arbor press to provide an even insertion.
Carbon paper and Dykem are common to analyze and refine interfering fit. Dykem comes in spray cans, Liquid with a brush in the lid markers and they have a spray removal solution. It sprays very thin if used properly.
This may be Your best bet to find out if the connectors are being poorly aligned, have too much pressure at specific points between the male and female connectors meaning the issue may be in manufacturing and not Your fault.
I suggested alignment pins for the opposite side of the connector and receiving holes would have to be drilled into the mating connector. These would have to be done by a very reliable machine shop.
In regards to; the existing pins may be to small allowing useless 'precision' due to the mating part rocking. Try a couple of descending thickness split shims to connect the terminals in stages, this will reduce some of the rocking. You may also try using an fixtured arbor press to provide an even insertion.
7
PCB/EDA/CAD / Re: Ltspice is a big dissapointment !
« Last post by sb4 on Today at 05:29:00 pm »I agree that the user interface is so non-standard that unless I use it daily, I have to relearn its quirks which takes time. Example: I wanted to do a quick sim of a simple circuit as a proof-of-concept. I first used the online version of Circuitlab which I've never used before. In about 10 minutes I had an answer. The interface is so intuitive that I was able to draw and simulate without referring to documentation. I think the software also has "smarts" that helps get to an answer.
I really liked the pallet of "reference parts" on the left, which seem to be standardized parts that are typical. One part I needed is a bridge rectifier module. I have not yet found such a part in LTspice as installed. It was easy to place parts, connect them, and run a basic transient simulation.
Because Circuitlab requires a subscription to save your circuits, I decided to do the same thing with LTSpice, which I used extensively years ago. Immediately I hit frustration as I could not remember the quirky GUI. I wanted to select a part to rotate it, but there is no generic select tool. You need to use the "move" hand-shaped cursor, and then press ctrl-R. I doubt I would have discovered that by trial and error, I had to google. The lack of a general select tool constantly stumps me. You have to use a "scissors" tool to delete a part, etc. I am also flummoxed by the lack of scroll bars on the schematic window. You can zoom in and out, but your schematic may drift out of view. Until I know better, I end up zooming out and trying to move the entire schematic into the center of the window, then zooming back in.
Selecting parts can be daunting since rarely used parts and popular parts are lumped into long lists. Very often we just want to grab a typical reference part with wide specs to check proof-of-concept of a circuit. There may be a way to set that up in LTspice, but it doesn't jump out at you. Circuitlab just has a nice common list of parts visible from the get-go.
I have a lot of confidence that LTspice simulations are well designed, fast, and accurate, though you may have to research how to tweak parameters for best performance. So I'm mainly complaining about the GUI. I'm aware that the guy who has been maintaining LTspice has left and now is maintainint QSpice, which is his complete rewrite of LTspice and supposedly has much better code. On first look, the GUI isn't any more friendly. However, perhaps now we can put in requests for improvement to the LTSpice GUI and get more response. An alternate GUI option like Circuitlab's would be my suggestion. Keep the original, but have an option to select the alternate.
I really liked the pallet of "reference parts" on the left, which seem to be standardized parts that are typical. One part I needed is a bridge rectifier module. I have not yet found such a part in LTspice as installed. It was easy to place parts, connect them, and run a basic transient simulation.
Because Circuitlab requires a subscription to save your circuits, I decided to do the same thing with LTSpice, which I used extensively years ago. Immediately I hit frustration as I could not remember the quirky GUI. I wanted to select a part to rotate it, but there is no generic select tool. You need to use the "move" hand-shaped cursor, and then press ctrl-R. I doubt I would have discovered that by trial and error, I had to google. The lack of a general select tool constantly stumps me. You have to use a "scissors" tool to delete a part, etc. I am also flummoxed by the lack of scroll bars on the schematic window. You can zoom in and out, but your schematic may drift out of view. Until I know better, I end up zooming out and trying to move the entire schematic into the center of the window, then zooming back in.
Selecting parts can be daunting since rarely used parts and popular parts are lumped into long lists. Very often we just want to grab a typical reference part with wide specs to check proof-of-concept of a circuit. There may be a way to set that up in LTspice, but it doesn't jump out at you. Circuitlab just has a nice common list of parts visible from the get-go.
I have a lot of confidence that LTspice simulations are well designed, fast, and accurate, though you may have to research how to tweak parameters for best performance. So I'm mainly complaining about the GUI. I'm aware that the guy who has been maintaining LTspice has left and now is maintainint QSpice, which is his complete rewrite of LTspice and supposedly has much better code. On first look, the GUI isn't any more friendly. However, perhaps now we can put in requests for improvement to the LTSpice GUI and get more response. An alternate GUI option like Circuitlab's would be my suggestion. Keep the original, but have an option to select the alternate.
8
Repair / MOVED: Looking for the pinout of the Li-Ion backup battery in HP Gen10 servers
« Last post by RoGeorge on Today at 05:21:13 pm »This topic has been moved to Projects, Designs, and Technical Stuff.
It started as a pinout request for a backup battery used in ProLiant HP servers, the info is nowere to be found, so it all turned into a HW+SW project about detecting the pinout + talking with a MAX17817 battery management chip over the SMBus
https://www.eevblog.com/forum/index.php?topic=426565.0
It started as a pinout request for a backup battery used in ProLiant HP servers, the info is nowere to be found, so it all turned into a HW+SW project about detecting the pinout + talking with a MAX17817 battery management chip over the SMBus
https://www.eevblog.com/forum/index.php?topic=426565.0
9
Microcontrollers / Re: Spurious pin-change interrupts with AVR DB
« Last post by pcprogrammer on Today at 05:16:35 pm »Sometimes it is not more capacitance, but less that does the job. For a linear regulator the datasheet specifies what capacitance is needed near to the regulator to avoid it from oscillating. Same applies to a lot of IC's where for instance a 100nF capacitor needs to be placed near to the supply pin.
I had problems with my MCU resetting due to voltage dropping to low when starting a brushed dc motor. Was using a buck converter to lower the voltage from 9 to 3.3 and the input voltage dropped to low also depleting the capacitors for the mcu. Solved it with a diode between the 9V and the buck converter and adding more capacitance for the mcu. That solved the turning on issue. To filter noise when stopping the motor I used ferrite beads and small capacitors plus series resistors and bigger capacitors to get a cleaner supply voltage.
Take a look at the schematic here.
Changing the polarity of the interrupt might also do the trick, but making sure the supply voltage is as clean as possible never hurts.
I had problems with my MCU resetting due to voltage dropping to low when starting a brushed dc motor. Was using a buck converter to lower the voltage from 9 to 3.3 and the input voltage dropped to low also depleting the capacitors for the mcu. Solved it with a diode between the 9V and the buck converter and adding more capacitance for the mcu. That solved the turning on issue. To filter noise when stopping the motor I used ferrite beads and small capacitors plus series resistors and bigger capacitors to get a cleaner supply voltage.
Take a look at the schematic here.
Changing the polarity of the interrupt might also do the trick, but making sure the supply voltage is as clean as possible never hurts.
10
Projects, Designs, and Technical Stuff / Re: Blinking Light pattern for maximum attention distance at night
« Last post by LaserSteve on Today at 05:16:15 pm »Day peak eye sensitivity is around 570 nm yellow. Night peak is around nm green.
Flashing is a must against urban background lighting. Daylight Warning strobes for radiotowers are 30-60 short [ xenon strobe ] flashes per minute, switching to short bursts of flashes at night at reduced power and reduced rate, often with a change of color to red. The night adapted brain is confused by a single flash but very much adapted to locating bursts of short flashes.
Aeronautical warning lights have always had a secondary use for navigation back to the 1930s, so they have far higher intensities then they actually need.
Alternating green and yellow or green and white single flashes makes for a strong means to locate a position, which is why that particular sequence is legally reserved for locating airports in the US.
200 to 1000 microsecond flashes l Xenon) [actually out to 100s of milliseconds] win, period. Running them in burst-pause-burst makes them distinctive
IALAs quick flashing and occulting patterns come in second.
Alternating the location of flashes aka wig-wag aircraft lighting comes in third.
Alternating color in short bursts comes in fourth.
The US FAA has this well spelled out for Anticollision, Tower, and Airport Beacon lighting in various regulations.
Recently in the US, alternating yellow green, of a pattern distinctive from the Aviation Beacon pattern has been adapted for construction vehicles on highways. in my State, Ohio, it was so effective on winter snowplows that a petition circulated to the governor's office to tame it by reducing intensity. It uses alternating colors and motion, and it is Highly effective.
https://youtu.be/X5G2Ya0ZaJ0?si=eN9nHL5FAubGEswb
https://youtu.be/ZAGZVblbS3U?si=S0ro6vgBL_ynvYyI
Ohio State trucks now use alternating yellow and white in the upper left and lower right and alternating green yellow in two other locations on the rear. Colors "flip" about twice per second. Using four spaced beacons with a quadrature phased [90 degree] color wig-wag sequence is amazingly effective.
State Law had to be amended to have green lighting on a vehicle, and requires it to be distinctive from traffic lights at intersections.
Do the pilots a favor and make your sequence different from Morse Code H, white yellow green, white green, white yellow, white white green, beacons.
Steve
Flashing is a must against urban background lighting. Daylight Warning strobes for radiotowers are 30-60 short [ xenon strobe ] flashes per minute, switching to short bursts of flashes at night at reduced power and reduced rate, often with a change of color to red. The night adapted brain is confused by a single flash but very much adapted to locating bursts of short flashes.
Aeronautical warning lights have always had a secondary use for navigation back to the 1930s, so they have far higher intensities then they actually need.
Alternating green and yellow or green and white single flashes makes for a strong means to locate a position, which is why that particular sequence is legally reserved for locating airports in the US.
200 to 1000 microsecond flashes l Xenon) [actually out to 100s of milliseconds] win, period. Running them in burst-pause-burst makes them distinctive
IALAs quick flashing and occulting patterns come in second.
Alternating the location of flashes aka wig-wag aircraft lighting comes in third.
Alternating color in short bursts comes in fourth.
The US FAA has this well spelled out for Anticollision, Tower, and Airport Beacon lighting in various regulations.
Recently in the US, alternating yellow green, of a pattern distinctive from the Aviation Beacon pattern has been adapted for construction vehicles on highways. in my State, Ohio, it was so effective on winter snowplows that a petition circulated to the governor's office to tame it by reducing intensity. It uses alternating colors and motion, and it is Highly effective.
https://youtu.be/X5G2Ya0ZaJ0?si=eN9nHL5FAubGEswb
https://youtu.be/ZAGZVblbS3U?si=S0ro6vgBL_ynvYyI
Ohio State trucks now use alternating yellow and white in the upper left and lower right and alternating green yellow in two other locations on the rear. Colors "flip" about twice per second. Using four spaced beacons with a quadrature phased [90 degree] color wig-wag sequence is amazingly effective.
State Law had to be amended to have green lighting on a vehicle, and requires it to be distinctive from traffic lights at intersections.
Do the pilots a favor and make your sequence different from Morse Code H, white yellow green, white green, white yellow, white white green, beacons.
Steve