Activity for TonyStewart
| Type | On... | Excerpt | Status | Date |
|---|---|---|---|---|
| Comment | Post #290874 |
FR-4 and soldermask will slowly absorb moisture under pressure. https://www.nextpcb.com/blog/waterproof-pcb--nextpcb (more) |
— | about 2 months ago |
| Comment | Post #290595 |
When I did Hipot testing, I modified the tester with a string of resistors to current limit and prevent destructive failures yet trigger insulation failures in the 100uA test. When I tested 5 MVA transformers to 200kV, I used an AM radio and LeCroy scope to measure partial discharge (PD) events befo... (more) |
— | 3 months ago |
| Comment | Post #290595 |
"When should a transformer core be potted?"
>When being attacked by environmental stress from all sides.
"Should it be potted at all? "
> It is not mandatory. Consider if Dk affected the frequency applied or the thermal temperature from insulation. This may not be in your case, but just ... (more) |
— | 3 months ago |
| Comment | Post #290720 |
My assumptions were based on analyzing the photo of the transparent substrate that the membrane switches were bonded to without any conductive traces. Yet on the same side below them was the carbon printed traces connecting the domed membranes beneath the clear substrate. It is true that membrane sw... (more) |
— | 3 months ago |
| Comment | Post #290448 |
Can you measure or compute the L & C values? https://www.ti.com/seclit/ml/slup376/slup376.pdf (more) |
— | 4 months ago |
| Comment | Post #290354 |
Charge pumps suffer from high very impedance output impedance leakage issues and are not worth the effort.
Murata package is my best recommendation.
https://www.digikey.ca/en/products/detail/murata-power-solutions-inc/NMV0512DC/5799183 (more) |
— | 5 months ago |
| Comment | Post #287196 |
I have successfully solved your problem in the past using a spreadsheet to to determine all the correct values for a potentiometer to give the the desired s21 response from 5% to 100% of the pot value chosen.
I trust you can easily do the same.
Anectdotal
__________
In my case I wanted to d... (more) |
— | 8 months ago |
| Comment | Post #287302 |
If "base voltage isn't applied" then Ib=0 & dIb=0 then the proposed formula with Ib is not logical. (more) |
— | 9 months ago |
| Comment | Post #288260 |
The slip is necessary for the rotor to generate the torque required for rotation. This force also draws more current. (more) |
— | 11 months ago |
| Comment | Post #288195 |
Otherwise using PWM pulsed energy to store power into a suitable inductor with CCM continuous drain you can match the average impedance with a circuit that does this with clamp protection where the % of drained current determines the reactive impedance as a voltage booster to a small stored capacito... (more) |
— | 11 months ago |
| Comment | Post #288195 |
Yes If 6V exceeds the reverse voltage rating on the PV then a diode is needed. When the voltage drops below 6V the energy is getting quite small % so harvesting under these conditions requires high impedance loads and there are custom IC's to do this with great effort for small reward.
Perhaps yo... (more) |
— | 11 months ago |
| Comment | Post #288225 |
Note: I just chose a small voltage with gain of -10 (more) |
— | 11 months ago |
| Comment | Post #288225 |
no. but you may want to define the purpose of the attenuator with a equation or function (more) |
— | 11 months ago |
| Comment | Post #288010 |
@Carl That's good. It doesn't have to be the right leg either. It could be the hip or some other body part that doesn't move much. Then you can see what you are doing
with the lights on ;) (more) |
— | 11 months ago |
| Comment | Post #287952 |
Low impedances suppress high impedance E fields so very high impedance inputs do not easily attenuate. Also very low magnetic B fields can induce low current which as you know if you have a nano-amp per meter induced on a one megohm R you get a 1mV signal which is a 1000 times bigger than a microv... (more) |
— | 11 months ago |
| Comment | Post #288184 |
@#60854 SInce "no-load= open circuit" test, we are talking about mechanical load to a motor with zero torque load or electrical - no load when power is cut off to measure back EMF. (more) |
— | 11 months ago |
| Comment | Post #288183 |
Ensure there are no stored voltages on any pin when power is applied, including avoiding switch bounce.
To test an IC with power off one can apply 2Vac with a 10k to 100k series resistor to each pin to check for short circuits or ESD damage. The V-I plots may be shown in XY mode on any scope an... (more) |
— | 11 months ago |
| Comment | Post #288010 |
A good layout is one that meets or exceeds your expectations (or design specs). These specs form what is called Design for Manufacturability (DFM) for Testability (DFT) for cost (DFC) and for performance of accepting signal inputs by quick connection, signal conditioning for gain, bandwidth and CMRR... (more) |
— | about 1 year ago |
| Comment | Post #286900 |
If I understand what you said refers to my description of a differential amp (DA) with 3 NPN's , " There is a negative feedback in this common-base (CB) stage driven by an input current " then I cannot agree.
The emitters are "common" to input and output so it is not a CB but rather differenti... (more) |
— | over 1 year ago |
| Comment | Post #287265 |
You would need a spectrum analyzer to determine the signal from line noise due to SMPS , motor starts and Triac dimmers in order to determine the Noise rejection required. But yes you should expect interference with the wider -40 and -60dB BW unless it was at least 6 to 8th order. (more) |
— | over 1 year ago |
| Comment | Post #287257 |
It seems you still use the single-ended topology but integrate current offset error with Hall, Rs or dual CT sensors to correct for the asymmetric current in each supply. (more) |
— | over 1 year ago |
| Comment | Post #286900 |
The simple ideas behind the differential amplifier are to ;
- eliminate the Vbe diode voltage offset from the input and choose any reference voltage for 0 output.
- use Ie to modulate gain when required
- choose either output for inverting or non-inverting gain or use both. (more) |
— | over 1 year ago |
| Comment | Post #287257 |
Does that mean you want it direct coupled but behave like a transformer AC coupled? Then the PFM or PWM drivers must have error feedback for the integral sum of current sensed from each to null the DC current and voltage feedback to null the error from the input voltage. So the answer lies in the ... (more) |
— | over 1 year ago |
| Comment | Post #285107 |
>Why does the collector current depend linearly on the base current?
When emitter resistance to supply rail is much greater than the base-emitter resistance re=26/Ie [mA] the voltage applied to the base is conducted to the base with Vbe drop. Thus there must be some relatively constant current g... (more) |
— | over 1 year ago |
| Comment | Post #287027 |
Conductance = S=siemens
(Volume) Conductivity, σ is "siemens per metre" (S/m) is the length/surface area
Resistivity (rho) = "ohm-metre" = area/ length (more) |
— | over 1 year ago |
| Comment | Post #286900 |
"Common" refers to input and output references that are shared. Vce and Vbe depend on the external resistors. While "e" is common to both.
(more) |
— | over 1 year ago |
| Comment | Post #286900 |
The differential BJT amplifier is well-defined in simpler terms.
Av=Rc/(re+Re)=Rc/(26/Ie+Re) (more) |
— | over 1 year ago |
| Comment | Post #285107 |
We know Ic is precisely controlled by the exponential function of Vbe and current gain is very inaccurate. This means every device has a wide range of Ic/Ib in the linear range and exceptionally wide variation in the nonlinear range down to 10% of the maximum current gain at Vce=Vce(sat) or usually r... (more) |
— | over 1 year ago |
| Comment | Post #286900 |
The common terminal is not the least important, perhaps it is often ignored how important it is. (more) |
— | over 1 year ago |
| Comment | Post #281971 |
I expect your measurements are accurate enough, but now you realize how reflections within 10 wavelengths or so can affect your results. (more) |
— | over 1 year ago |
| Comment | Post #286950 |
A bipolar switch rated for high Vceo seems to be all you need to protect the driver. Either high side PNP or low side NPN. >100V is easy to find. Your 1st plot is not to scale. (more) |
— | over 1 year ago |
| Comment | Post #286944 |
The ideal conditions are almost never achieved in ordinary ground communications, due to obstructions, reflections from buildings, and most importantly reflections from the ground out of phase. This fading can cancel up to all of the signal. So it is more important you define the path and all refle... (more) |
— | over 1 year ago |
| Comment | Post #286215 |
"across it" or it's own voltage is merely a value indicating the state of charge at some time. The actual charge Q = CV(t) for the capacitor's voltage. Your question is just about the semantics of a single test point (with an assumed common test point of 0V) or differential "across"
(more) |
— | over 1 year ago |
| Comment | Post #281971 |
What have you learnt? (more) |
— | over 1 year ago |
| Comment | Post #286868 |
interesting analysis +1, close enuf for gov't work. (more) |
— | over 1 year ago |
| Comment | Post #282053 |
@#8176 The reality is your results of variations are real and certainly affected up to 10 or more wavelengths from reflections.
The other reality is s21 is also affected by these reflections with multipath, so the net effect is to measure RSSI where possible in your receiver and test the variation... (more) |
— | over 1 year ago |
| Comment | Post #286211 |
Anti-aliasing filters are always needed when the spectrum above the Nyquist rate fs/2 will degrade SNR with the resulting difference frequency output. Thus dynamic range and spectral values ought to be estimated with desired SNR.
Typically a sampling rate might be 3x the upper frequency is the sp... (more) |
— | over 1 year ago |
| Comment | Post #286215 |
**Is this interesting?**
All voltages are relative to some reference. The local ground is defined as zero volts, just like Protective Earth PE ground. So if one side of the cap. is 0V then the voltage drop is the same as "the voltage" (relative to 0V.)
**What is reactance, really?**
It is ... (more) |
— | over 1 year ago |
| Comment | Post #286862 |
Thanks Olin, I added details to explain for others to understand why it is 180 + 180, with the assumption he knows there exists a frequency point at which -180 deg occurs from 3rd order (x90deg) phase shift) from 2L's and 1 C in a Pi filter. This is a slight change, from the Hartley. (more) |
— | over 1 year ago |
| Comment | Post #286859 |
Not precisely, but you can filter the harmonic content of a broadband pulse to pass a narrow BW filter. Or you can phase modulate a square wave and filter any harmonic or fundamental.
(more) |
— | over 1 year ago |
| Comment | Post #286209 |
Level shifting can be accomodated with >10k resistors to limit ESD diode current rating specified.
(more) |
— | about 2 years ago |
| Comment | Post #286209 |
IF you want to detect tri-states, you need to specify the condition for the high-Zstate such as back driving with a 10k resistor 1/0 and detect that. (more) |
— | about 2 years ago |
| Comment | Post #281444 |
But we only wore tinfoil hats when I bought a Lingren EMI proof chamber. In the 90’s. (more) |
— | almost 3 years ago |
| Comment | Post #279493 |
I remember when Qmodem would compute the bit rate differently, and shown higher than output expected. On dialup using 56k modems in high speed mode with a faster serial port, they would count the start and stop bits in the transfer rate, which then of course got translated into , I forget 64 levels ... (more) |
— | almost 3 years ago |
| Comment | Post #279493 |
Can you give an example where Baud Rate is higher than Bit Rate? (more) |
— | over 3 years ago |
| Comment | Post #279493 |
... So don't byte me for saying your Byte rate example is not related to bit rate vs baud rate or symbol rate. but does affect character rate. characters are not symbols in this sense yet are symbols in language. (more) |
— | over 3 years ago |
| Comment | Post #279493 |
There are many types of Baud Symbols that compress bits in time, frequency, phase or time in many combinations by sending many per unit of a bit. such as 9600 modems using 1200+2400 Hz tones with multiple levels to get 8 bits per Baud symbol in a 2400~3000 Hz bandwidth. ( I think, I forget)
(more) |
— | over 3 years ago |
| Comment | Post #279493 |
Dear Olin, I think you are using bit rate efficiency with overhead to define effective data rate, both which have nothing to do with the bit/s baud compression technology
Symbols can stuff more bits per symbol such as the Baud symbol in a modem after Émile Baudot who invented the Baudot code (Bodo... (more) |
— | over 3 years ago |
| Comment | Post #279010 |
The cap must be low ESR type as there is also an R ratio from arc negative resistance to cap ESR voltage divider. The. Common mode series R and TVS diodes provide secondary clamping with current limiting R chosen for 1kV to TVS Rs. The arc -ve Rs was proven by Faraday to be inverse to current densi... (more) |
— | over 3 years ago |
| Comment | Post #279010 |
Erhm When the resistor arcs it will become a capacitance transformer, 200pF/100nF = 0.2% of 15kV
so move the caps to the shunt R and make the current loop small and close to the injection PT. The cap must be low ESR type as there is also an R ratio from arc negative resistance to cap ESR voltage di... (more) |
— | over 3 years ago |