开云体育

I ordered one

but no schematic??

--
Jack, W8TEE

Sorry! Wrong link!

The right one is

http://www.nd6t.com/uBITX/AGC.htm

On Tue, Sep 21, 2021 at 03:15 PM, Tom.VA7TQB wrote:

Is it possible to get a schematic of the board?
Would like to understand how it interfaces with the existing circuit.
(bought the board a year ago and just about to install it)

73 de PU1OWL

Rafael

Hello:

Consider me old fashioned or just plain dumb?

Is it possible to get a schematic of the board?
Would like to understand how it interfaces with the existing circuit.
(bought the board a year ago and just about to install it)

?Thanks
? ? Tom
? ? VA7TQB

Finally had time to get back to this for the second time. I accidentally deleted the first post I made of this a week ago.
Here is another set of test results with the AGC board vol-hi connected to to TP21 on the uBITX via a resistor.
(remember this is a V4 board)

I made three sets of measurements
1. AGC off
2. AGC-slow and a resistor value of 11.5 ohms
3. AGC-slow and a resistor value of 5.75 ohms

11.5 ohms is the closest I had to the 10 ohms Tim suggested.
Hoping to get a little higher levels for the S-Meter a 5.75 ohms (2x11.5 in parallel) was tested to see if it would increase the bias voltage and in turn the ADC reading.

Interestingly the measurements was almost exactly the same with both resistors. There looked like a very slight instability on the oscope with 5.75 ohms. after the fact that may have been due to a poor connection. In any case I'm going to stick with the 11.5 ohm resistor

From the measurements its pretty clear the AGC is doing its job and is pretty effective for about 50db. It looks like it kicks in by S9+10
and continues to work to at least S9+50. By S9+60 it has given up. Maybe someone else can use the measurements to come up with some useful numbers.

Bottom line is -- connecting the AGC board to TP21 through a resistor of about 10 ohms works for me and I'm a happy clam, I mean ham. I would though love to see some one else replicate these tests.

73 digger AB3XU

Ooops, accidentally deleted my last post. Will repost tomorrow.

Yes! That's exactly why I meant. The signal present at the s-meter pad is the function of some math based on the time constants of the resistor switched in and the 1uF Capacitor in the circuit. So yeah it will be different with different amounts of modulation.

The Bias voltage for the S-Meter is what controls the amount of attenuation the board presents to the incoming signal. So the fact that its clamping a S9+40 to a "comforatable" level in comparison to a S3 or so means it 100% is working as expected! Which is awesome news. If you want some more gain you can probably drop that 10Ω resistor down to like 5Ω. But with out some testing/math I do not know if that will saturate the audio pre-amp. I will look into this once I get a chance!

I am pretty sure that's the range that my device exhibits as well. I very much believe that it is divided by 4 to fit it in a byte as well. The configuration manager says as much with the range being 0-255 instead of 0-1024 as would be expected by the ADC.

-- Tim Keller - KE2GKB

On Tue, May 7, 2019 at 06:48 PM, Adrian Chadd wrote:

It's likely divided by four so the s curve entries can be bytes (1024/4 = 256), rather than 10 bit entries.

Makes sense. Thanks Chadd

Tim, I can definitely see it respond on the oscope and the audio volume stays pretty consistent until about S9+40, maybe +50. I'll take? better look in the next day or two.

By density, do you mean some actual signal to modulate the carrier? I have a two-tone oscillator kit I haven't gotten around to building yet. I could put some test setup together with it, the KX-3 and some attenuators I don't yet have.? Qualitative measurements are nice, but I like real measurements better.

If the bias is controlling the gain then the board is doing it's job as intended. But as far as drive for the S-Meter the voltage I get is a bit low.? At S9 it is only 84 out of 1024 as measured by the raduino's ADC, not a lot of resolution then between S0 and S9 especially since the CEC firmware (for some reason) divides that by 4. I'll have to look at why Dr. Lee's doing that.

-- digger AB3XU

That actually has a much more appropriate response curve! I feel like the issue now is more related to signal density then anything else. As in with that 10Ω Resistor we are finally getting an appropriate response :)

How does it perform as an AGC in that wireup? I guess that is the "real" test anyway.

-- Tim Keller - KE2GKB

Mark,
I'm hardly an expert at what I'm doing so, I can't be too far ahead of you.

With your signal generator and voltmeter you ought to be able to make the same measurements I have. Give us some idea of what first baby steps need help with. I don't know the HP8640b but I can try to help with where to connect and take measurements.

-- digger AB3XU?

Tim,
Had a chance last night to make some more measurements. Hopefully we are the same page now.

I made two sets of measurements to look at the difference between taking the audio on the ubitx on either side of R70 -- vol-h and TP21. Also curious I made them with USB and LSB. So.....

Sark 110 (7.101Mhz) --> ubitx (7.099.7 USB)
???? vol-h --> AGC, vol-h
???? TP21 --> 11.4ohm --> AGC, vol-h

Sark 110 (7.100Mhz) --> ubitx (7.100.7 LSB)
???? vol-h --> AGC, vol-h
???? TP21 --> 11.4ohm --> AGC, vol-h

That tuning on the ubitx gets a tone very close to 1K. (i guess that means I need to adjust the BFO by about 300Hz)
The resister is 11.4 ohms - my other choice was 9ohms.
AGC was in the middle/slow position.

The measurements I'm getting with using TP21 for the input to the AGC still seem lower by about half than what is expected based on ND6T's circuit theory and numbers he posted in another thread.

ADC Comparison of ND6T's numbers and my readings from CEC firmware
RF???????? ND6T??? mine
S9?????????? 170???????? 84
S9+10??? ?? -????????? 170
S9+20???? 300??????? 200
S9+30?????? -?????????? 297
S9+40???? 700??????? 399
S9+50???? 800??????? 517

I hope this all makes sense! Let me know if anything need interpretation!
I would really like to see some measurements made with another V4.

Sounds like "Digger" (AB3XU) is a few notches more experienced. However, I do have a HP 8640b, HP652A and HP 3400A that I can apply on the issue if necessary. Just will need "baby step" instructions.?

Be happy to help. Your project kits are well appreciated!

73

Mark
AJ6CU

Correct! Though it may be more then a week or 2 still, My day job has been leaving me less time then desired to work on this. that aside I want to believe that not all v4's have this issue for as many AGC's as we have sold we have gotten maybe 8-12 of you that have spoken out about issues with a v4 board. That being said even with the "Issues" the radio still works as expected the AGC action is just lack luster compared to what it should be!

If you want to verify it on your board you can run the +12V,GND, and audio in lines. Wire up the switch put it in either on position and measure the Voltage on S-Meter pad. if you get a decent voltage on a S9+ station odds are it will work fine for you. Or just wait if you dont trust your self to test it this way the boards can be finicky to unsolder due to the nature of castellations. So use your judgement based on your skill!

-- Tim Keller - KE2GKB

Tim,

Just confirming since I also have your AGC and a v4 board....? Sounds like you are close to a solution to support v4 boards but may be a week or so to confirm?? Installing the AGC was next on my mod list, but I can defer it for a couple weeks and go chase some other gremlin.? :-)

73
Mark
AJ6CU

No worries! Its all part of the learning process for all of us :) And nothing astray, it's really hard to articulate exactly ones intentions over text, But we have been making progress in some form or another!

Ooops! No worries having it wrong would have just put the signal at 1KHZ in LSB instead of USB. That being said and knowing what you are measuring it all matches what I expected! Now that I know what the data is trying to show me.

The difference is going to be more of a how much current can be sourced. I believe that the issue is the 1k resistor(R70) is presenting to much impedance that it is failing to provide the current needed to drive the amplifier on the AGC. By bypassing R70 we remove that limitation. With that there really should be a 10Ω Resistor in front of the AGC if you are bypassing R70 for safety.

I have been looking for the measurement on A7 in reference to Vol-H. Vol-H should be taken with the AGC in the off position. However your signal generator numbers are more then sufficient in that regards. :) So sadly more tests to be done! For consistency sake can we have the AGC in "Slow" mode, This is the mode with the 1MΩ resistor switched in.

I need to actually do a comparison with my IC-718, The antenna is currently wrapped up due to some life issues....maybe ill get a chance this week. I'll report back later this week with comparison! I am genuinely curious my self.

-- Tim Keller - KE2GKB

Tim, do you have any other "commercial" rig with an s-meter? I'd be curious to know what your voltage measurements of 0.3V to 2V might read on another rigs s-meter.