What errors? I speculate and guess a lot (it's part of having ideas) but we design very high performance gear, and everything works rev A, without prototypes. I certainly don't put that level of discipline into newsgroup posts, because nobody pays me here.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
No, this is not a crystal radio. It is a Q measurement circuit. I've suggested more traditional ideal diode circuits for this, but AMDX wants to use the transistor and diode approach.
-- Rick C
The OP doesn't need to measure low levels. It is a Q measurement intended to measure the amplitude of a coil/cap resonance. The input can be adjusted to put the output at a suitable range. In fact, he is trying to improve the tube circuit in a Boonton 260A so he can measure Q values of over 500. I don't believe small signals at this point are a problem.
What's wrong with the old standby precision rectifier this way?
R1 R2 ___ ___ o---|___|---o-------|___|-----------. | | | | o------->|-----. | | | | | | | | |\ | | '-------|-\ | | | >---o--->|---o----o .-------|+/ | | |/ | | | .-. .-. | |R3 | | RL | | | | '-' '-' | | | | GND GND
(created by AACircuit v1.28.7 beta 10/23/16
Figure 15
-- Rick C
Not an inverted triode, a simple cathode follower which minimizes the capacitive load on the input. It needs no voltage gain and with a bias that barely drives any current through the cathode it acts as a rectifier (albeit a poor one).
He was considering a FET with a voltage divider on the drain, but now the circuit will use a high input impedance/low capacitance amp (FET source follower with a bipolar emitter follower driving various nodes through capacitors to mitigate input capacitance) driving a standard common emitter gain stage. This circuit was largely hashed out in another thread.
-- Rick C
since the OPs stated purpose is repairing a Boonton 260, I'd forget all about building a better diode detector and instead take it up a notch and consider using one of the LT or other compay's fine IC RF detectors.
They will give better linearity and temp stability over a wider RF level range then you will ever achieve with a diode.
m
I offered to send you a dollar a day, as long as you keep posting. :^) (I love your stuff!)
George h.
You need to talk to the OP, not me. I'm pushing to not use the simple diode circuit and he is not confident with using an op amp for a 10 MHz signal, so it is an uphill discussion.
If you want to propose a different circuit you need to provide him with more info than "LT makes some fine parts". I am no RF expert and AMDX is a total novice (except he is learning fast and is willing to get his hands dirty).
-- Rick C
A dollar a day? That changes everything.
I am sort of an idea factory, a consequence of having a schitzotypic personality. Some ideas are goofy and some are, after analysis, wrong. We analyze the hell out of any ideas that we advance on, so it's very rare that a bad idea makes it onto a PC board. Our mistakes are more pedestrian.
The schitzotypic thing isn't a personality defect, it's an asset. And to some extent, it can be taught to others. Electronic *design* needs ideas.
I'm dissapointed in how few people actually understand circuits, and how few people have ideas. I suppose I shouldn't complain.
Hey, the way to eliminate diode nonlinearity in an AM detector is to drive the diodes from a current source.
Or...
-- John Larkin Highland Technology, Inc lunatic fringe electronics
A diff pair with a current mirror load, driving a half-wave voltage doubler (the one with capacitor input) makes a pretty good AM detector. It has slew rate issues, but much less serious ones than an op amp "perfect" rectifier.
Cheers
Phil Hobbs
Even better with CMOS... I can do well under 1mV Peak, no-sweat (assuming no trim ;-) ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Thinking outside the box... producing elegant solutions.
Precision rectifier requires the op-amp to be faster than the signal. This circuit just requires the op-amp to be faster than you need updates.
-- Tim Wescott Control systems, embedded software and circuit design I'm looking for work! See my website if you're interested http://www.wescottdesign.com
The amp needs to keep up with the RF. So for low RF frequencies, you need a crazy-fast amp. And for high RF frequencies, there is no amp that will do.
-- Tim Wescott Control systems, embedded software and circuit design I'm looking for work! See my website if you're interested http://www.wescottdesign.com
Pretty much what I was thinking, an AC current source driving a diode doubler. I was thinking about a fast NPN on the low side, with a biggish emitter resistor, collector driving a PNP pseudo-inductor current source. AC couple that mess into the diode doubler.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Rick, sorry I wasn't mode clear. my previous response about IC detector alternatives was not directed at you specifically. M
Current source, or put a rectifier inside the feedback loop?
Both ideas are shown in this extract from an Norwegian ham experimenter:
piglet
That is the first RF detector I have seen that says it will work over
50kHz. I finally built a detector that worked and was linear, problem is it only worked around 1kHz. The detector you posted is now second on my list. I want to give James Arthur's circuit a second try. I could not get the first attempt to work.This what James posted.
****************************************************** Who needs self-biasing for an active detector?A fast peak detector might be done something like this:
Vcc Vcc -+- -+- | | |/Q1 |/ Q2 > --+--[R1]--| .--[R2]--+--| MMBTH10 | |>. | | |>. | | | C1 --- | '--------- | -' --- | | | | | === .-. | + - | | .------+------(M)----->| | VR1 22K | | '-' C302 --- [22k] R301 |
100nF --- | | | | | === === ===Wide dynamic range, and temperature-stable, too.
**************************************************** Any reason I shouldn't capacitive couple to the input? James, does C1 go down on the wiper of the pot? I built the first version with 2N4144 transistors, Could that cause any malfunction? What is the input voltage range? I'm using a 1200 ohm resistor to represent my meter, then monitoring the output with my multimeter. Could that 1200 ohms cause it to malfunction? Frustrated, Mikek-- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus
"Crazy fast"? It has a completely linear response and needs to handle
10 MHz for the OP. I think that is doable. No?How much non-linearity do you think your circuit will have? It might not be too bad if the leak resistor is large (the one you omitted). Without a leak resistor your circuit is not a rectifier, but a peak detector and won't respond when the amplitude is decreased.
With your circuit I would say an RC of ~50 Hz would work for the OP and still handle all frequencies he is interested in.
-- Rick C
It needs the d.c. voltage level from the prior stage. But I just simulated it and it's amazingly awful, I'm not sure why.
The MPSH10 is a 2GHz ft RF transistor. I'm not familiar with a 2n4414, but yes, the transistor matters quite a lot.
We'd need to know your meter's deflection sensitivity to size it to your Boonton's meter. I was posting the idea-- the topology--not a finished, scaled circuit.
My detector's a bust, but this is a better (faster) front end, substituting a fast RF transistor for the emitter-follower. I show it running on B+ for fun, with ZD1 for safety, just in case.
Since you actually want a pF or so of capacitance to match the tube, we don't need to bootstrap the drain.
I used a less component-sensitive bias scheme, but you still might have to customize the bias to accommodate your particular FET.
+250V -+- | R4 [150k] 1/2W | +25V .---------------+-----+------+------. | | | | | R1 [10M] | | C4 --- ZD1'-. | T1 |--' |100n --- 20V ^ 4.6V | 2n4416 | | | || | | | | R3 [20M] | | | | C3 | | | | 100pF | |/ Q2 +------+---||---+---| MMBTH10 | | |>. R2 [2.2M] R5 [1.5k] | | | | === | | 6.2V '-----+--------> to detector -1dB @ 50MHz | Rin > 500M @ 5MHz R6 [3.3k] Cin = 0.6pF @ T1(g) | ===
Sorry about the detector, but it sure sucks. I'm missing something.
Cheers, James Arthur
I understand, but if you want to help the OP he will need a lot more guidance. I'm not building a circuit. He is.
-- Rick C
That makes sense. They make some darn fast opamps these days. At a max frequency of 10MHz... sorta ~10ns "edges". (period/10) I've never used an opamp rectifier that fast so there's likely to be a lot of gotcha's.
George H.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.