Re: Confused about how AGC works?

Jerry Gaffke

A poorly designed AGC circuit could oscillate exactly as you describe.
Whether or not it will oscillate depends on the delays through each element
and the gain of each element.  Both can vary depending on the frequency
at which they are calculated.
https://en.wikipedia.org/wiki/Barkhausen_stability_criterion

Jerry, KE7ER

On Thu, Apr 18, 2019 at 06:00 PM, digger AB3XU wrote:
To begin with I can follow along with the theory and it does makes sense. Very simply, the signal from Vol-Hi via Q1 is used to control the conduction of Q2 and how much Q3 shunts the IF signal. Ideally RF levels below S9 are left relatively untouched, however if greater than that, the action of Q2 conducting less and Q3 acting as a shunt  results in reducing clamping (reducing) the IF level.

So here's where I get lost.  As the IF level gets attenuated by the action of the AGC doesn't that in turn:
1. attenuate the level you will see at Vol-HI
2. which will then decreases the effect of the AGC
3. leading to an increased IF signal level
4. and increasing Vol-HI
5. which then increases the AGC's effect
6. reducing the IF signal level
7. resulting in a feedback loop of the IF level going up and down...
Am I describing my confusion/question in a way that makes sense?

73, digger AB3XU

Re: Trial to control CW power (and reduce harmonics as a result) by unbalancing the 2nd mixer

John (vk2eta)

Thanks Jerry,

Yes that gives me good reference points. So assuming the clocks #1 and #2 are at the same level, it looks like after this mod the signal at T2 (3,5) could be around 10 to 16dB too high.

I like that QRPGuys probe.

All the best,

73, John

Re: Trial to control CW power (and reduce harmonics as a result) by unbalancing the 2nd mixer

Jerry Gaffke

Signal level into the D1,2 mixer can be a bit higher than it is into the D3,4 mixer
when transmitting SSB.  But not by much.

Assume the D3,4 mixer has a signal loss of 6dB, the 45mhz crystal has a loss of 4dB,
and that the 2n3904 based v3,v4 uBitx 45mhz IF amp has a gain of 10dB.
(Gain is less than the designed for 16dB because the 2n3904's
can't give that much gain at 45mhz.  I have no idea what v5 does.)
We wind up with the same level of SSB signal into the D1,2 mixer
as we had going into the D3,4 mixer.

On Thu, Apr 18, 2019 at 08:15 PM, John (vk2eta) wrote:
Yes that gives me good reference points. So assuming the clocks #1 and #2 are at the same level, it looks like after this mod the signal at T2 (3,5) could be around 10 to 16dB too high.

Re: Trial to control CW power (and reduce harmonics as a result) by unbalancing the 2nd mixer

Jerry Gaffke

I like the display on the QRP Guys RF probe.
However, the input to the AD8307 is not well protected,
and any DC at the probe tip will blow the diodes.

Here's a cheaper AD8307 RF probe with what looks like a better front end, but no display:
Cheaper still, but without the better front end:

On Thu, Apr 18, 2019 at 07:03 PM, Jerry Gaffke wrote:
I'm liking the looks of this AD8307 RF probe, very clever how they make use of a cheap digital voltmeter module:
https://qrpguys.com/qrpguys-digital-field-strength-meter

Re: Trial to control CW power (and reduce harmonics as a result) by unbalancing the 2nd mixer

Joe Puma

I found another kit and info for the AD8307

Joe
KD2NFC

On Apr 19, 2019, at 12:08 AM, Jerry Gaffke via Groups.Io <jgaffke@...> wrote:

I like the display on the QRP Guys RF probe.
However, the input to the AD8307 is not well protected,
and any DC at the probe tip will blow the diodes.

Here's a cheaper AD8307 RF probe with what looks like a better front end, but no display:
Cheaper still, but without the better front end:

On Thu, Apr 18, 2019 at 07:03 PM, Jerry Gaffke wrote:
I'm liking the looks of this AD8307 RF probe, very clever how they make use of a cheap digital voltmeter module:
https://qrpguys.com/qrpguys-digital-field-strength-meter

Re: next ubitx

Ashhar Farhan

There are several ways to do this. ALC is just one way. An easier approach is to do it in software. This needs a software that can control the 'mic volume'. you could set the value differently for each band.

there is another pay-off with software mic gain, it can make a major difference to the tx IMD. At voice peaks, the tx linear chain compresses. The gain is not constant between low and high levels of modulation. This is what the in-channel IMD is. Now, if we have a look up table that amplifies the peaks more than the lows, we can 'correct' the gain back to being linear.  This simple concept goes by the name of 'pre-distortion' in the SDR world.

-f

On Thu, Apr 18, 2019 at 9:21 AM Jerry Gaffke via Groups.Io <jgaffke=yahoo.com@groups.io> wrote:
If you want to electronically adjust TX power out on the uBitx, talk to John, VK2ETA

He's got that working with no hardware mods for SSB:
https://groups.io/g/BITX20/message/45781
and now with some minor mods for CW:
https://groups.io/g/BITX20/message/67981

Jerry, KE7ER

Honestly though, if we had an ALC input to drive low to decrease TX gain and thus output, we could create our own power control loopbacks and SWR protection. It'd make 100W operation less.. crazy.

(I'd like to do this, but I'm still diagnosing the basics on the v4 board; I'd like the TX drive out to be clean before i feed it into an amp and waste more power..)

No Boot-Up (after spikes)

Ted

It came to pass that,  in the process of inserting a 7809 between the Raduino daugterboard and its B+ pin-to-mainboard, that I failed to notice a solder blob shorting out the 7809. I made a 90-degree header change at the Raduino mounting point on this build, making this relatively painless [or not].

It was realised, when turning on the (1-metre-long) remote power switch whereupon an added power relay chattered on and off [due to low relay supply voltage for the instant of the short],  graciously saving my 5-amp fuse.   Don't ask....

Since then, the radio powers up but the Nextion screen would not get past simply powering on and showing the basic screen graphics (as if in the PC editor). The obvious have been performed:

- Replace Nano.
- Try different brand of Nano
- Try Nextion screen from redundant uBitx radio
- Replace Raduino (in stock, known)
- Replace 5-volt regulator on oruginal Raduino.
- Try swapped Raduino with different Nano
- Restore Raduino with original, 2-line LCD display (shows opening call sign and firmware edition, nothing more)

So: should not a Raduino board power up by itself when given 12v & (-) , and respond to the original LCD display or a Nextion screen, that is, respond to manual display inputs even without a mainboard present?

If a mainboard is required for Rauino/screen operation past initial lighting, what might fail specifically which will stall further brain function like that?

V3 board; whole rig seen in my Photos folder.

73,

Ted
K3RTA

Re: Attenuator function in CEC firmware

Tom, wb6b

On Thu, Apr 18, 2019 at 04:26 PM, Jerry Gaffke wrote:
You could easily control that attenuation from a processor, perhaps
These may be a little overboard for the purpose but I have one of each of these digitally controlled attenuators that I bought to build up my tool set for making RF measurements.

This one had the part number of the attenuator chip ground off. Depending on the manufacture of the part, the specified low frequency cutoff could be 9Khz, 1Mhz or 10Mhz. For HF measurements a possible 10Mhz cutoff is an issue.
https://www.ebay.com/itm/New-OLED-6G-Digital-Programmable-Attenuator-30DB-step-0-25-RF-Module-Micro-USB-/263584061842
However, the construction with the machined aluminum body is impressive.

This one specifies the part number in the description and the number on the attenuator chip matches. It is a 9Khz to 4Ghz attenuator. It doesn't have a display, so would be a better choice for installing into the uBitx.
https://www.ebay.com/itm/PE4302-Digital-RF-Step-Attenuator-Module-High-Linearity-0-5dB-50-ohm-RF-DSA/121830578745

You can also buy the current version of the Peregrine attenuator chip from Digikey, mount it on a SMD adaptor board and add it to your uBitx IF chain. But, the modules are easier to work with.

These parts have an actual 50 ohm impedance (when the other end is 50 ohm terminated). It might be that the 50 ohms is a bit more load that the IF chain was designed for.

Tom, wb6b

Re: No Boot-Up (after spikes)

John (vk2eta)

Hello Ted,

Sorry about the mishap, but it has happened to quite a few of us before.

I would try the Raduino off the main board but connected to a PC for supply via USB.

If it is gets stuck at firmware version I would suspect either a defective si5351 clock generator or an issue with the i2c bus (A4/A5 from memory). Check for shorts, and if you have a scope check that you get a 3.3v swing on the two i2c signal wires.

73, John

Re: No Boot-Up (after spikes)

Ted

Thanks for the tip, John.  Thing is that I've tried no fewer than four Nano replacements and two other, known operational Raduino boards.  That's the part that makes no sense. Bad outputs or clock readings should have been eliminated by now due to component-swapping.

Tnx agn,

Ted
K3RTA

Re: No Boot-Up (after spikes)

Sam Tedesco

Nextion shows callsign/ver and all zeros on freq?

Re: Attenuator function in CEC firmware

Jack Brabham - KZ5A

Tom,

The 4302 or the newer 4312 (which seems hard to find) are the modules I'm thinking of.    I think the 4302 was mentioned in ZL1AXG's article.

The placement between the LPF and the pre-driver puts the ATTN in the antenna path on RX which is presumably 50 ohm and I assume is also 50 ohm on TX considering the by-directional LPF.

This placement, to non-engineer me, looks like the ideal "effector" location for an RX AGC scheme, ahead of any active devices.    Or it could just be setup as a manual RF gain control on RX.   It is also looks like a very straight forward way to control drive levels per band on TX, either via pre-sets or an ALC setup.

I'm also thinking that this approach should be largely free of "unintended consequences".

In any case it looks like an enjoyable experiment.

73 Jack KZ5A

On 4/19/2019 5:10 AM, Tom, wb6b wrote:
On Thu, Apr 18, 2019 at 04:26 PM, Jerry Gaffke wrote:
You could easily control that attenuation from a processor, perhaps
These may be a little overboard for the purpose but I have one of each of these digitally controlled attenuators that I bought to build up my tool set for making RF measurements.

This one had the part number of the attenuator chip ground off. Depending on the manufacture of the part, the specified low frequency cutoff could be 9Khz, 1Mhz or 10Mhz. For HF measurements a possible 10Mhz cutoff is an issue.
https://www.ebay.com/itm/New-OLED-6G-Digital-Programmable-Attenuator-30DB-step-0-25-RF-Module-Micro-USB-/263584061842
However, the construction with the machined aluminum body is impressive.

This one specifies the part number in the description and the number on the attenuator chip matches. It is a 9Khz to 4Ghz attenuator. It doesn't have a display, so would be a better choice for installing into the uBitx.
https://www.ebay.com/itm/PE4302-Digital-RF-Step-Attenuator-Module-High-Linearity-0-5dB-50-ohm-RF-DSA/121830578745

You can also buy the current version of the Peregrine attenuator chip from Digikey, mount it on a SMD adaptor board and add it to your uBitx IF chain. But, the modules are easier to work with.

These parts have an actual 50 ohm impedance (when the other end is 50 ohm terminated). It might be that the 50 ohms is a bit more load that the IF chain was designed for.

Tom, wb6b

Re: No Boot-Up (after spikes)

Joe Puma

It will do that if one of the yellow or blue wires are not connected forgot which one. Probably the RX line. So maybe that connect is damaged at the Raduino.

Joe

On Apr 19, 2019, at 9:10 AM, Sam Tedesco <stedesco619@...> wrote:

Nextion shows callsign/ver and all zeros on freq?

New V5 build results and calibration issues.

derek (G4VWI)

The following observations may be of interest to others having just got round to completing my V5 build.

1) For safety I made use of the SMT reverse polarity protection circuit by KC9ON to simplify the power supply fuse circuitry which worked very well. Two separate fuses used.

2) I  used the Kit-Projects version of the ND6T gain control. Out of three versions used this one is by far the easier to fit and is recommended as it has a 5v regulator working from the 12v line. I have used a tiny mil. spec. rotary switch instead of the DP3T switch provided. Adjust for a little more gain to allow for compensation when switching in a narrow CW filter.
3) I have used Murata T filters with .01 caps to ground in the power lines. I was getting noise from a small brushless fan. If anyone has used a SMALL temperature switch circuit let me know. If not where have others wired their fans to? I am not keen to solder directly onto the IC.
Initial tests:
1) Out of the box the VFO was out by 1.2Khz! Others mention this but as yet no explanation has been given. To calibrate make sure you have a dummy load connected. Go into "settings" >"calibrate". Connect the PTT orange wire to ground and a carrier should be visible on your analyser, SDR or second radio display. Mine was set to under 1K bandwidth centered on 10Mhz. and no trace was visible! Widening the bandwidth still no trace until the relay was activated for the third time. Possibly a sticky relay I hope? First attempt I ended up with the tuning high. Second attempt zooming in it was near enough. The BFO calibration agreed with other published settings and was straight foward.
2) Checking the CW output on a spectrum scope  showed the resistor values were not ideal and the keying will need adjusting when CEC firmware is installed. IAMBIC B produced better results. Suggestions?
On 20Mtrs. a burst of RF appeared after the keying had completed. I am not sure if this was due to the exposed circuitry on the bench or if this is a known issue with a solution? Let me know.
3)
Checking the power output the meter (not calibrated) showed an average between 5-8 watts depending on band.
4) This V5 example appears to be much more sensitive than my V4.
5) The LM386 audio amp gives adequate output into a small Visaton FRWS-5 4 Ohm speaker.
NB: The PA's get quite warm to the touch with key down even with a larger heat sink so a small fan is advised.

I intend to add a Nextion 3.2e display. Will the saved parameters and calibration be read when updating the firmware? Has anyone added a RTC to the display yet?

Please let me know how you have got on with your V5 build and comment on any points raised.

Best wishes. Derek G4VWI.

Re: Confused about how AGC works?

Don, ND6T

Hi Digger,

Often the simplest of things are revealed as complex. As Jerry says, the delays are important. The values used allows the attenuation to smoothly increase and track the signal level variations such that the response is fast enough to produce a pleasant result but not so fast that it will oscillate. Your confusion may be that you perceive that the end result is a tightly controlled output. It is not. Stronger signals result in slightly stronger audio levels. A S9+40 dB signal will be a bit louder than a S9 signal, just not horribly so. That increase is designed to be slight and gradual. This simple circuit is almost (but not quite) linear in that respect. Fortuitously, the shunt resistance takes over as the series resistance nears maximum conduction and it is a nice, gentle, transition. Close but not perfect. Good enough.

To get a better AGC, especially one that tames the initial burst, you would need a much more complex circuit. The better ones include an additional detector and multiple additional gain control points throughout the RF and IF amplifiers. If we had a carrier to use as a signal strength reference then it would be easier. Without the carrier we have to rely on the modulated signal. Tricky.

I use a compromise circuit. Even then, it seems that it is so complex that the majority of builders need kits or completed boards. This circuit is the best that I can do (so far) in keeping it simple yet effective.

Did this help?
73, Don

Re: Attenuator function in CEC firmware

Tom, wb6b

On Fri, Apr 19, 2019 at 06:10 AM, Jack Brabham - KZ5A wrote:
looks like the ideal "effector" location for an RX AGC scheme
Hi,

Yes, that looks like it could be a good choice. Just be certain that the power level at that point, on transmit, is within the limits of the attenuator. Off hand I think it is, but there many folks here that may have a more accurate assessment. There have been some issues with Q90 being blown by nearby transmitters and lightning strikes. It may be a good idea to add one of the protection mods, like diodes, that have been suggested in other threads, to protect the attenuator chip input.

I may do the same with my attenuator module, enhance my uBitx, rather than my original reason for buying is as a piece of test equipment. I'll watch your progress.

Tom, wb6b

Re: Confused about how AGC works?

Jerry Gaffke

Here's my shot at a quick summary regarding AGC circuits for the Bitx40 and uBitx
Opinions may vary.

When tuning across the band with a stock radio, you might run across a weak station
and turn the volume up to hear it.  Then a local QRO station responds and
you have to rip the headphones off to protect your ears, or be very quick in reaching
for that volume control.  An AGC circuit (automatic gain control)  turns the volume down
for you when a strong station shows up like that.

This is an early version of Don's (ND6T) AGC circuit:
https://groups.io/g/BITX20/message/68037
Still viable, works well enough, simple and easy to build from scratch.
Really strong local signals can still be annoyingly loud, though much less so.

Don refined that FET shunt attenuator design by adding a FET series attenuator
to get additional dynamic range, keeps even the strongest stations down in volume.
http://www.nd6t.com/uBITX/AGC.htm
It is a bit more complicated, but performs better.
Was kitted up by Kees (no longer available),
and is now available from  https://shop.kit-projects.com/

Either of the above AGC circuits will work on the Bitx40 or the uBitx.

Don's design attenuates the RF signal as it comes in from the antenna,
using a control signal derived from the audio.

Most of the other AGC designs presented in this forum (there have been a
half dozen or so) attenuate the audio after the audio pre-amp, simple and
works well enough to protect the ears, and you avoid the tricky business of
attenuating the RF path.  Here's a very early example of this approach:
http://bitxhacks.blogspot.com/2016/11/agc-for-bitx.html

Attenuating the RF path is preferred, since the stock receiver dynamic range
is limited primarily by that audio pre-amp, and also to some extent by the IF amps
and mixers.  Don's design reduces the signal right where it comes in from the
antenna jack, so nothing in the radio will get overloaded and cause distortion.

So an AGC design with attenuation in the audio stage works to protect the ears,
but strong signals may be distorted enough to be unintelligible.  With RF
attenuation, strong signals remain undistorted.

All successful AGC circuit designs presented in the forum thus far detect the
incoming signal level just before the volume control.  Signal detection must occur
after the 12mhz crystal filter, otherwise we would also detect strong nearby stations
that the crystal filter rejects.  Detection of audio is not quite ideal since it takes time for
the signal to arrive there, so there will be a brief pop when a strong signal shows up
before it is properly attenuated.  But detection of 12mhz RF between the crystal filter and the
demodulator is difficult, because the unshielded 12mhz BFO on CLK0 will sneak into
our AGC signal detector and look like a very strong signal.

Here's an example of a well regarded AGC scheme for a high performance receiver:
http://www.ka7exm.net/hycas/hycas_200712_qst.pdf

Jerry, KE7ER

On Thu, Apr 18, 2019 at 06:43 PM, Curt wrote:
check this out - this is a similar version of the same circuit

http://www.nd6t.com/bitx/AGC

the main idea is sampling the audio, feeding the transistor to boost this signal.  the 2 diodes develop a rectified signal based upon the audio, and this is used to bias a MOSFET used as a variable resistor.

Re: No Boot-Up (after spikes)

Ted

That's that's about the only thing that's left course, and it may just be a case of denial that I would bother to check them. Guess I'll get on that.

Tnx,

Ted

Re: No Boot-Up (after spikes)

Ted

Sam,

Yes.  See earlier response. I thought I had wired up a virgin screen cord but it was late.   it must have been, because some other time during the night I destroyed the 2.8" screen that goes in this project. So a new one, the enhanced model this time, is somewhere between who knows where and where I am, in the mail. Let's hope i don't blow up the redundant screen while trying to troubleshoot, and I'll re-do those wires.  If that doesn't work, it's haunted,  and I have plenty of gas for chainsaw and a box of 20 gauge shells.

Ted

Re: Confused about how AGC works?

As discussed before…

“An AGC circuit (automatic gain control)  turns the volume down for you when a strong station shows up like that.”

That’s known as AVC or “Automatic Volume Control”… not necessarily AGC.

MRM

From: BITX20@groups.io [mailto:BITX20@groups.io] On Behalf Of Jerry Gaffke via Groups.Io
Sent: Friday, April 19, 2019 11:14 AM
To: BITX20@groups.io
Subject: Re: [BITX20] Confused about how AGC works?

Here's my shot at a quick summary regarding AGC circuits for the Bitx40 and uBitx
Opinions may vary.

When tuning across the band with a stock radio, you might run across a weak station
and turn the volume up to hear it.  Then a local QRO station responds and
you have to rip the headphones off to protect your ears, or be very quick in reaching
for that volume control.  An AGC circuit (automatic gain control)  turns the volume down
for you when a strong station shows up like that.

This is an early version of Don's (ND6T) AGC circuit:
https://groups.io/g/BITX20/message/68037
Still viable, works well enough, simple and easy to build from scratch.
Really strong local signals can still be annoyingly loud, though much less so.

Don refined that FET shunt attenuator design by adding a FET series attenuator
to get additional dynamic range, keeps even the strongest stations down in volume.
http://www.nd6t.com/uBITX/AGC.htm
It is a bit more complicated, but performs better.
Was kitted up by Kees (no longer available),
and is now available from  https://shop.kit-projects.com/

Either of the above AGC circuits will work on the Bitx40 or the uBitx.

Don's design attenuates the RF signal as it comes in from the antenna,
using a control signal derived from the audio.

Most of the other AGC designs presented in this forum (there have been a
half dozen or so) attenuate the audio after the audio pre-amp, simple and
works well enough to protect the ears, and you avoid the tricky business of
attenuating the RF path.  Here's a very early example of this approach:
http://bitxhacks.blogspot.com/2016/11/agc-for-bitx.html

Attenuating the RF path is preferred, since the stock receiver dynamic range
is limited primarily by that audio pre-amp, and also to some extent by the IF amps
and mixers.  Don's design reduces the signal right where it comes in from the
antenna jack, so nothing in the radio will get overloaded and cause distortion.

So an AGC design with attenuation in the audio stage works to protect the ears,
but strong signals may be distorted enough to be unintelligible.  With RF
attenuation, strong signals remain undistorted.

All successful AGC circuit designs presented in the forum thus far detect the
incoming signal level just before the volume control.  Signal detection must occur
after the 12mhz crystal filter, otherwise we would also detect strong nearby stations
that the crystal filter rejects.  Detection of audio is not quite ideal since it takes time for
the signal to arrive there, so there will be a brief pop when a strong signal shows up
before it is properly attenuated.  But detection of 12mhz RF between the crystal filter and the
demodulator is difficult, because the unshielded 12mhz BFO on CLK0 will sneak into
our AGC signal detector and look like a very strong signal.

Here's an example of a well regarded AGC scheme for a high performance receiver:
http://www.ka7exm.net/hycas/hycas_200712_qst.pdf

Jerry, KE7ER

On Thu, Apr 18, 2019 at 06:43 PM, Curt wrote:

check this out - this is a similar version of the same circuit

http://www.nd6t.com/bitx/AGC

the main idea is sampling the audio, feeding the transistor to boost this signal.  the 2 diodes develop a rectified signal based upon the audio, and this is used to bias a MOSFET used as a variable resistor.

 Virus-free. www.avg.com

--

…_. _._