|
|
LogProbe with a pick-up coil plug
LogProbe is a logarithmic detector that converts input signals to a DC output proportional to the logarithm of the input signal strength. If you are not an engineer this may sound complicated, but is the same way many of our senses works!
The full Moon illuminates us with about 0.2 Lux and full Sun light is 100,000 Lux. A power range of 1:500,000! Although we can surely perceive the difference, it does not seem like that much! A 57 dB range of power.
Or consider the difference in sound pressure from a quiet room (40 dB) to a noisy ship building facility (100 dB)! A difference of 60 dB or 1 million times in power!
Even the forces we can generate by means of our muscles are kind of logarithmic! We can hold a strand of hair between our teeth, and yet we can bite off a piece of tough bread with the same. Imagine a watch maker who also could have boxing or wrestling for a hobby! In the watch he (or she, these days) has to control the forces of hands and fingers to some 10 mN (the weight of one gram, although gram is a mass). In the boxing ring the forces may be 1 kN (100 kg or 200+ punds.) A range of 1:100,000 or 50 dB.
As you have figured out by now, the dB is the number of zeros of the ratio, times ten. 100,000 = 5 zeros, 50 dB.
This puts the phenomenal dynamic range (92 dB) of LogProbe in perspective. Add a very good accuracy to this!
|
|
|
The LogProbe Frequency Response
|
Response with frequency as the parameter. You can see how the first 7 frequencies, listed on the right, make up virtually one line. At 700 MHz and above it starts to drop off, but the slope is still very accurately a 10 dB/Volt output signal. |
Response with power level as the parameter. You can see how the lines are almost horizontal up to 600 MHz. A 0.3 V droop is apparent, making for a frequency dependency of approx. 0.5 dB/100 MHz. Above 700 MHz it is 6 dB/100. |
The degraded performance >600 MHz is most likely a somewhat temporary adapter for BNC connectors I had made. A better one is designed, and when built, I will repeat the measurement above.
The available generator had 1300 MHz as maximum frequency and +13 dBm max power, so the graphs are limited by this. Since these measurements the DC output level has been increased slightly. The devices are now adjusted to 5 V out for -30 dBm @ 100 MHz. This leaves a "tail" for no input signal of ~ 0.5 V or less, allowing for signals down to some -90 dBm to at least be detected, if not yet on the "dB linear" part. The smallest detectable signal (single ended mode, across 50 Ohm) is ~20 µV.
The inputs on LogProbe are truly differential. For single ended operation, one can be grounded. Signal on either input results in a positive going output signal. More about how to treat the inputs further down on this page!
The output signal spans typically 0.1 V to 9.3 V, for the 92 dB range of 55µV<Vin<2.2V on the input. This represents -75 to +17 dBm across 50 W and -83 to +9 dBm across 600 W. For cable TV, using 'dBmV' in 75 W , it is -25 to 67 dBmV.
Output linearity is better than ±1dB and the effects of temperature over -40<T<+85 °C are on the order of ±0.1 dB/K.
In designing LogProbe the concept was to create a detector that is inexpensive but sensitive and sturdy enough to be used throughout a laboratory, to be light and small enough to stick in the pocket and take out on a job, and to be accurate and reliable enough for to be regarded as a real instrument. Not just 'a detector'.
For these reasons the hardware is based upon the DIN standard audio connectors for which plugs, receptacles and shielded extension cables are readily available. This allows for the customer to put together his (or her) own measurement heads, systems or accessories.
LogProbe has about 20 different heads and accessories. Some are under development and more to come.
|
|
Note: For signal out on pin 5 a simple addition of an internal wire is required.
|
Front view and pin numbering, their functions and connections to the leads.
The inputs:
Normally you only have to get a differential signal to pins 1 & 3. They 'float' at about 3.6 V above ground as from a ~ 4kW source. The operation is unimpeded if they are forced to between 1.7<Vin<5. If they are forced to 0<Vin<1.7 the operation seizes, but LogProbe is unharmed. Forcing the inputs outside 0<Vin<5 is will destroy the LogProbe. So, use insulated wire, or if you are using a blank wire loop or coil, put a Teflon sleeve on it! The thickness of the wire must be 1.10 ~ 1.25 mm to fit the socket. This corresponds to 43 ~ 50 mils. AWG #17 falls in this area and #16 would work, but is a bit on the thick side. If you need to ground the center (or one end) of the loop or coil, you have to do so via a capacitor to pin 2.
There is no '+' or '-' input since a signal to either, with respect to the other, will contribute to a positive going output signal. For low frequencies, the output voltage corresponds to the logarithm of the momentary input signal. A triangular waveform will then look almost like a full wave rectified sine wave on the output. Above ~10 MHz the variations are too rapid and the output signal assumes a DC voltage, corresponding to the log ('dB') of the input level. For to read low frequency signals' levels on a meter or a scope, it may be necessary to low pass filter the (frequency doubled) output signal. A meter movement is a low pass filter in itself, (especially since the movement needs a resistor in series anyway, so an electrolytic capacitor can be connected across the movement itself) allowing for audio signals from 20 Hz (giving a 40 Hz ripple on the output) to be read directly.
If you for example are testing the noise on the output of a DC power supply at >5 V, be careful! You can use electrolytic capacitor(s), one to ground or the - side, and one to the + side of the supply. Turn up the supply slowly, so the capacitors can charge up via the internal ~4 kOhm without ripping the inputs outside the 0 - 5 V range. AC noise can now be observed. Be careful not to short the power supply since it would be the same as turning down the voltage too fast! You will need some 3 µF to see 50 Hz. The input impedance is 1.1 kW || 4.2 pF (differentially) when a DIN plug is used.
A better, and safer, method for higher voltages would be to make a voltage divider with resistors, bringing down the voltage to a safe level! Or add clamp diodes from each input to ground and a +5 V source.
Log probe is supplied with a few 'blank' DIN connectors for the customers convenience. When a blanks is used, the added capacitance is ~1 pF.
Sometimes it is desirable to use a pre-amplifier, a buffer or some other active device in the probe head. LogProbe is designed with this in mind! V+ is what you supply to the red wire (within the allowed 3<V+<35) and it is available on pin 4. Actually, you can feed V+ into pin 4 if it is more practical, leaving the end of the red wire open. Pin 2, GROUND, is then the supply return, as well as the green and the black wires. Output is on the yellow wire. The output can easily be wired internally to pin 5, in which case the pin 3 is the best candidate for grounding if LogProbe is used single mode.
The power consumption is a mere 20 mA typ. LogProbe works at 3V already, but the output of 100 mV/dB is of course limited to <3 V or a ~ -50dBm reading then. With 5 V operation the maximum output is of course limited to 5 V and if the demand on output current is heavy you will get even less. The LogProbe has an internal 50 W back termination in the output line, delivering 100 mV/dB in an open circuit, 50 mV/dB into a 50 W load and 2 mA/dB into a short. The max current drive is ±>25 mA with unlimited capacitive load.
The rise and fall times in response to a pulsed RF source depends some on the signal amplitude. This will be investigated further, but ~ 1µs is in the ball park.
BATTERY INCLUDED!
A 9V battery is supplied so the eager owner can use LogProbe right out of the box!
"Shipping and Handling" nonsense is not my style. I pay for the shipping within the US. $250 covers all.
California residents have to add 7.25% sales tax. For air mail outside the US, add $10!
By the time this battery is out, you may have located a wall plug adapter, or bought our WP-115 or WP-230. Any other DC power supply or battery of 5.5<V<32 will do. At V>24 the maximum output current may not be permanently available due to overheating of the internal 5V regulator (=> thermal shutdown). If the regulator shuts down, from overheating, no permanent damage is done. It will start up again as it cools off. At room temperature about 10 mA is permanently available at V+ = 35V. There is no particular reason to run LogProbe on more than 10 V for most applications, and at 12<V<15 for the occasions where an attached head requires that voltage. The remaining range from 15 to 32 V is available for your convenience
THE LogProbe IS NOT YET PROTECTED AGAINST REVERSE POLARITY!
(This may come as a re-design)
It has been determined, by destructive experiments, that a reverse voltage is not destructive if it is limited to about 20 mA. It is unlikely that a power supply is, and a battery is certainly not, so take great care! You reverse the power and blow the unit, you will have to buy a new one! Even a current limited reversed supply will be able of a short pulse of excessive current and will not do the Tantalum capacitor good anyway in the long run.
THERE IS NO INPUT PROTECTION FOR INPUT VOLTAGES OUTSIDE 0<Vin<5V EITHER!
Protection would seriously affect the sensitivity, so I have opted for not to include it.
Although I guarantee the products to work as advertised, I cannot replace units that have been destroyed by excess input voltages or reversed/excess supply voltage. I reserve the right to determine the reason for failure on returned units.
For further questions, mail or orders, please contact us at
