EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Lazor on March 31, 2021, 02:43:27 pm
-
Hello,
I have build HV probe 1:8000
https://www.youtube.com/watch?v=ZJHYIH9O-JU (https://www.youtube.com/watch?v=ZJHYIH9O-JU)
I have plan to measure it with SDG2042X and SDS2202X-E to draw bode plot, but there I have few questions:
1. Input of probe is Hi-Z (200M \$\Omega\$ probably 250fF) Should I add 50 \$\Omega\$ terminator parallel to probe input?
2. I wonder if I can skip 50 \$\Omega\$ load and just do measure on Hi-Z. Cables that I planned to use is RG316 50cm, max frequency 15MHz. I don't know if could I have problem with signal reflection with it?
3. Is it possible to measure 1mV with SDS2202X-E? In manual they write lowest 500uV/div but it is only manual, they can write everything ;)
Thanks
-
15 MHz has free space wavelength of 20 meters. Your 30 cm cable length can be considered negligible for most purposes.
-
Hello,
I have build HV probe 1:8000
https://www.youtube.com/watch?v=ZJHYIH9O-JU (https://www.youtube.com/watch?v=ZJHYIH9O-JU)
I have plan to measure it with SDG2042X and SDS2202X-E to draw bode plot, but there I have few questions:
1. Input of probe is Hi-Z (200M \$\Omega\$ probably 250fF) Should I add 50 \$\Omega\$ terminator parallel to probe input?
2. I wonder if I can skip 50 \$\Omega\$ load and just do measure on Hi-Z. Cables that I planned to use is RG316 50cm, max frequency 15MHz. I don't know if could I have problem with signal reflection with it?
3. Is it possible to measure 1mV with SDS2202X-E? In manual they write lowest 500uV/div but it is only manual, they can write everything ;)
Thanks
Welcome to the forum.
Oh shit scary stuff !
Be careful with those sorts of voltages !
1. For Bode plot with SDS2202X-E you do have the option to select 50 Ohms for channel input impedance however take care not to exceed the input rating when you use it.
2. If you're having problems with HF reflections selecting 50 Ohm input mode can provide a better transmission line match.
3. Certainly and even the 500uV/div is real and certainly handy when using 100:1 or 1000:1 probes for their voltage ratings but needing to resolve small amplitude signals.
-
1. Input of probe is Hi-Z (200M \$\Omega\$ probably 250fF) Should I add 50 \$\Omega\$ terminator parallel to probe input?
No, such high voltage generator is unable to keep high voltage on a 50 Ω load. If you're planning to use it with generator which can keep 100 kV at 50 Ω load, this is very dangerous. Because such voltage generator needs to provide 100 MW power into your 50 Ω terminator. It will leads to a big bada-bum and a large plasma ball :-BROKE
2. I wonder if I can skip 50 \$\Omega\$ load and just do measure on Hi-Z. Cables that I planned to use is RG316 50cm, max frequency 15MHz. I don't know if could I have problem with signal reflection with it?
You can skip it on the input. But if you're planning to measure 15 MHz, you're needs pass-through 50 Ω terminator at oscilloscope connector and your voltage divider should be calculated for 50 Ω load on the output.
If you're planning to use high impedance probe to measure the output of divider, then there is no needs for 50 Ω terminators. But then you're needs to take into account that your divider will be loaded with high impedance load. It will affect voltage on the output.
Also note, that your voltage divider has parasitic inductance and capacitance. From the photo it looks like a series of coil inductors made from resistors, such inductance may affect amplitude and waveform at high frequency.
Yes, it will have wave reflection, it will happens in your voltage divider, due to specific geometry. But since you're planning to use low frequency, I think it shouldn't affect your measurement much.
-
One can measure without 50 ohms termination at the generator side, as the cable is short. Even with not so short a cable with good match at the generator not much happen - except the doubling of the amplitude. A higher amplitude at the input to the divider is more like a good thing.
The critical case to avoid is a "long" cable with poor termination on both sides.
The cable from the probe to the scope is more critical - try to keep it as close as possible to the later use case.
-
1. Input of probe is Hi-Z (200M \$\Omega\$ probably 250fF) Should I add 50 \$\Omega\$ terminator parallel to probe input?
No, such high voltage generator is unable to keep high voltage on a 50 Ω load. If you're planning to use it with generator which can keep 100 kV at 50 Ω load, this is very dangerous. Because such voltage generator needs to provide 100 MW power into your 50 Ω terminator. It will leads to a big bada-bum and a large plasma ball :-BROKE
There is missunderstand. I have plan to use 20Vpp from SDG2042X. I plan use 100kV to check voltage rating but without 50 \$\Omega\$ load ;)
And I miss information that this is active probe.
15 MHz has free space wavelength of 20 meters. Your 30 cm cable length can be considered negligible for most purposes.
I find this page(1):
https://www.switchcraft.org/learning/2016/12/9/challenges-using-pwm-in-motor-drive-applications (https://www.switchcraft.org/learning/2016/12/9/challenges-using-pwm-in-motor-drive-applications)
and this(2):
https://studylib.net/doc/18203254/calculating-the-propagation-delay-of-coaxial-cable (https://studylib.net/doc/18203254/calculating-the-propagation-delay-of-coaxial-cable)
If I have 15MHz then rise time is approximately 1/3 of period (based on https://design.iconnect007.media/index.php/article/59799/rise-time-vs-frequency-whats-the-relationship/59802/?skin=design (https://design.iconnect007.media/index.php/article/59799/rise-time-vs-frequency-whats-the-relationship/59802/?skin=design))
so 15MHz rise time is ~20ns and propagation time for RG316 is 1.464ns/ft. For 50cm propagation delay is ~2,4ns. Now based on equation 12 from first link. Based on it I got voltage reflected to generator termianal is ~27,2Vpp. This is correct? and if it is correct it could damage front end of generator?
The critical case to avoid is a "long" cable with poor termination on both sides.
"long" cable is depended on frequency how I guess. I don't know if I can damage generator when output of generator is connected to HiZ HV probe input. I have calculate reflected voltage value, but I don't know if it could damage generator?
Thanks all for response
-
Th function generator is OK with a open cable - the 100 M load would not make things worse.
Fast edges from a square wave include more than 15 MHz, but the measurement of the frequency response up to 15 MHz should still be OK. The function generator should have a reasonable termination at the output - so even the higher frequencies that are reflected would be absorbed in the termination and would not cause resonances.
An open output can be a problem to higher power RF amplifiers, not made for this. There the output termination is usually not from an actual resistor at the output, but by the output impedance of the actual amplifier.