![/!\](/moin_static199/modern/img/alert.png "/!\"){kind=small}
Vorläufig / Preliminary, work in progress
Active RF Probe
On ebay.de (Oct. 2017) you can find a low cost (about 13 EUR) Active RF Probe from company transverters-store as a module only. You have to take care about the cable to the oscilloscope, the power supply, and a housing/shield.
Pros:
- low cost
- comes with documentation
Cons:
- There is some ringing on a square wave signal
you can magnify the pictures (and download) with a mouse click on it.
|
Specifications
- Input device - BF-998
- Input Impedance - 0.75 pF / 10 MOhm
- Bandwidth - 100 kHz to 1.5 GHz (± 2.5 dB)
- Gain - –20 dB nominal
|
- 1-dB compression point - not measured
- maximum input voltage up to 250 V
- Power supply 7 - 12 V @ about 35 mA
- Size - 55 x 8 mm
Ground lead, flexible, length = 5 cm (recommended)
Coax cable: 1 m RG-58 with BNC connector, diameter 5 mm (recommended)
|
I am using a 9 V rechargeable battery as the power supply. That should last for about 5 hours use.
Schematic
I have drawn my own schematic, so you do not have to search for.
For easier printing use the .pdf file.
In case you want to make changes to the schematic, you can download the Eagle 7.5 file.
The part numbers do not conform to the part numbers of the original schematic.
|
Housing / Shield
If you apply no housing / shield you can see a 400 mVpp 50 Hz signal.
I wanted a standard part for the housing. So I came to an aluminum tube from a hardware store. See the picture/drawing on the right.
The board size has a width of about 8.2 mm. So, it does not directly fit into the tube. So, I drilled with 8.5 mm 48 mm deep, to make the board fit into the tube. It has also the benefit, that you have a defined stop for inserting the PCB. For grounding I drilled a hole M3 into the tube and made a cutout to access the coax cable shield. In the cutout area I removed the coax cable isolation to have the blank shield. A M3 soldering eyelet made then the contact from the housing to the coax cable shield. At last solder the eyelet to the coax cable shield.
I soldered at the end of the Ground lead an IC socket pin, the same can also be used as a connector for Ground at the measurement place.
|
Signal Quality
For a test I used a Square Wave source (10 MHz, CMOS, 5 V). See here for details.
Unfortunately there is some ringing (about 120 MHz) on the signal, compared with a Hantek PP-90 probe (bandwidth 80 MHz), see the screen shot on the right.
See in the Links a similar project for an Active Probe. Here the author recommends a 33 Ohm resistor in series with the output, to the coax cable. But in my case this made the ringing worse.
The signal shows the best shape, if there is a 50 Ohm termination at the oscilloscope input. But that lowers the amplitude level.
That can be compensated with a replacement of the first of the three (in series) 2.2 pF input capacitors with a 4.7 pf capacitor. That increases the input capacity from 0.75 pF to 0.9 pF. See the first picture above for the place of the capacitor.
|
In the picture to the right you see my probe connection to the 10 MHz calibrator, in order to have the same test point for both probes.
Ferrite Bead
In the left screen shot below you see a ringing of about 30 mVpp (30% of the signal).
In order to lower the ringing, the Calibration oscillator got one Ferrite Bead inserted (5 mm long, 3,5 diameter):
in the signal output line.
That reduces the ringing a little at the source by about 17% (30 mVpp to 25 mVpp). So, it can be considered as an option.
See the screen shot on the right below, with oscilloscope Tektronix 475 (bandwidth 200 MHz), and probe RS489-706 (bandwidth 250 MHz). This is my best result in measuring the 10 MHz square wave oscillator.
|
|
|
Trial: Tip Shield
Because of the high impedance and a non negligible length of the measurement tip, it is sensible for RF noise. See those Technical notes for a more detailed explanation. Tom K1TRB helped me a lot on this topic.
The capacity from tip to ground (without shield) is 2 pF. With shield it is 32 pF. So, the tip shield does not make too much sense. It is better to measure in a shielded and grounded box (environment).
So, following this Construction plan I used the Aluminum tape, which I used before for the Yana project. You get it from ebay under the name "48mm x 17m Roll Silver Aluminium Foil Adhesive Repairs Heating Duct Sealing Tape", order number 142017488471, for about 2.70 EUR.
The measuring tip is spring-loaded. So, the fixed part should be isolated, preferably with shrink tube. See the picture on the right. That allows to apply the shield very close to the moveable part of the measuring tip.
|
Next you cut a piece 3 x 2 cm of the conducting Aluminium foil. Recurve it about 5 mm to the lower side. That part must be brought in contact with the ground contact area. The rest is wrapped around the board, the lower end must complete with the end of the fixed part of the measuring tip.
See the picture on the right.
That end should be fixed with a piece of shrink tube.
A few spectrum screen shots will show you the result of the effort.
When you live in a city area, there is a lot of RF noise, as you will see.
On the left picture below you see the spectrum of the Active Probe with an unshielded tip. You will notice a few Radio bands.
On the right picture below you see the spectrum of the Active Probe with the shielded tip, but with no power supply. I assume that the coax cable has a too weak shield attenuation (about 38 dB).
|
|
On the left picture below you see the spectrum of the Active Probe with the shielded tip and power supply. It shows very little difference to the previous picture. For me, that is a good sign, that the shield works efficiently.
On the right picture below you see the spectrum of the Active Probe with the shielded tip and power supply. In this case I hold my finger at the tip, to simulate an antenna. The Radio bands are clearly visible.
|
|
Oscilloscope Tektronix 475
Because of the high ringing amplitude of the modified Hantek oscilloscope, another Tektronix 475 (Bandwidth = 200 MHz, input capacity 20 pF) was used.
Next the Active RF probe was testet with the same signal as before. In the left screen shot below with a ground lead of 8 cm (3 inch) length, and in the right screen shot with a ground lead of 5 cm (2 inch) length. That shorted ground lead reduces further the amplitude of the ringing.
|
|
|
If you do not terminate the Active RF Probe coax cable (RG-58, 1 m long) at the oscilloscope input with 50 Ohm, the screen shot on the right shows the disturbed signal.
Summary
After a long trial and error time, I think this Active RF Probe is usable for sensible RF measurements. Working with a 10 MHz square wave generator (rise time < 6 ns) is very tricky. The most important thing is a good ground contact beside the point to measure.
Links
Contact Email
Please enter your Email address, if you expect an answer.
The entered Email address will not be published, or given away.
List of pages in this category:
-- RudolfReuter 2017-10-25 20:40:26
Go back to CategoryTechDoc or StartSeite