EMRFD Message Archive 15264

Message Date From Subject
15264 2019-04-13 09:43:02 Mike Hadley HyCas

I now have only three complete 9.000/12.000 Meg HyCas kits left for sale. After these have gone boards only will be available.

Check out my information about my Product detector and audio board.

Contact me by email on mhadley157@gmail.com for further details.

Mike G4JXX

Sent from my Xperia™ by Sony smartphone

15265 2019-04-13 14:56:15 Ronald Strathman Re: HyCas
I'll take another Hycas kit. Maybe as a spare or use in another project. Don't recall the price.
Have you any experience with the PHSNA project? 
Ron  K6DNV/XE2

Virus-free. www.avast.com

15266 2019-04-13 16:04:02 mhadley157@gmail.... Re: HyCas
Hi Ron.

Sorry but the response to the last available HyCas kits was very quick and now they have all been sold.

I do however have plenty of pcbs. If I look hard enough I may be able to cobble together another kit for you or if there is still interest from others in full kits I will consider putting together another batch.

I now have a working prototype S meter board with a 10 segment led bargraph display specifically designed for the HyCas. Again, if there is enough interest I will make a pcb available.

Regarding the phsna project I'm afraid I have not had any dealings with it but it looks a quite exciting (and cheap) way to build a very useful piece of kit.

For more info on kits and boards email me direct. mhadley157@gmail.com

Mike G4JXX

Sent from my ASUS

-------- Original Message --------
15288 2019-05-25 09:21:44 bwbangerter Re: HyCas
I recently assembled one of the HYCas kits offered by Mike, G4JXX. All the parts were provided and correct, and assembly went well. The last couple of days I have been characterizing the circuit, my tools including a homebrew DSS signal generator, commercial step attenuator, homebrew AD8307 rf power meter, and 100 MHz Tektronix 'scope. The board is powered from a 12V supply, and the output of the 9V regulator is 8.94V. A 5K potentiometer is installed at R32 as the IF gain control.

I see the following with no rf input to the circuit: When the Mute input is high, the AGC line voltage is 0.63V, a diode voltage drop as would be expected from the schematic, regardless of the IF gain setting and regardless of whether the AGC is on or off. When the Mute input is off, the AGC line voltage varies from 1.01V to 7.16V as the IF gain control is rotated from ccw to cw. So far so good.

With rf input at 9.00 MHz between 0 dBm and -67 dBm, and the IF gain setting at maximum ( full cw), and the AGC on, the rf output of the circuit varies from -32.5 dBm to -33.2 dBm, very “tight” AGC action indeed, and this level is adjusted with R6, the Threshold Set control (500 Ohm pot), which could just as accurately be called the output level control. The AGC line voltage increases from 2.66V to 6.00V as the input is varied from 0 to -67 dBm. For input levels of -68 dBm and below, the AGC line voltage is 7.16V, so the AGC threshold is -67 dBm, and the output drops in accord with the decreasing input level.

With the AGC off, the gain if the circuit is ~38 dB for input level increasing from -70 dBm to -50 dBm, the output increasing from -32 dBm to -13 dBm. As the input level is increased above -50 dBm, the gain drops off, distortion of the waveform sets in, and a maximum output of ~ -6 dBm (and highly distorted) is reached.

The original W7ZOI/WA7MLH paper reports an AGC OFF gain of 55 to 60 dB, whereas I see 38 dB. An AGC threshold of -95 dBm is claimed (Application Hints addendum from the W7ZOI web site) whereas I see a threshold of -67 dBm. An AGC range of 100 dB is claimed. I see considerably less than that, though since the AGC line voltage is 2.66V at 0 dBm input, there would appear to be greater gain reduction to be obtained in going to higher input levels, and I have not exceeded 0 dBm. I expect my build of the HyCas circuit would far more nearly approximate that of the original design in the “open loop” gain were 20 dB higher.

Does anyone with experience with this circuit have suggestions? Note that Rin, the first stage gate to ground resistance is 3.3K and L1 is 6.8 uH, though it ought to be closer to 7.1 uH for an optimum input match. C2 does peak the gain nicely, however. Also, the gate to ground resistors for stages 2 and 3 are 1K. Presumably increasing one or both of these values would increase the open loop gain. My circuit is stable, and has shown no tendency to oscillate.

Ben, K0IKR
15322 2019-05-30 10:23:12 jim_amos_n8cah Re: HyCas
Hi Ben,

I'll measure my HyCAS with regards to AGC Activation and maximum gain.   I'll do a plot of Pin, Pout, and VAGC to see how it responds at different levels.  I had similar issues with my HyCAS after I built it as well and also suspected that I wasn't see the maximum gain.  In my case, I was working with a fairly rough PCB that had a lot of errors.  So, at first I wasn't sure if it was a circuit error or some other issue.  I think I eventually worked out all of the circuit errors. 

I have been using mine on a 20/40 meter receiver successfully.  I don't have the ability to switch it to manual gain easily though, I just leave AGC on.   I have it hard wired for maximum gain and have been primarily experimenting with crystal filters with it.    But one thing I do need to do (which might also effect maximum gain) is to put it into a metal enclosure.   I can also easily see de-sense from my BFO depending on the filter that I'm trying.   


15328 2019-06-03 09:24:08 bwbangerter Re: HyCas
Thanks, Jim. I look forward to seeing your results. Steve, AA7U reports observing a gain of 50 dB, an AGC threshold of -80 dBm with -35 dBm output level. He has a Roger Hayward kit with R7=3.0K, R1=1.0K, R2=2.2K.

15329 2019-06-03 15:26:49 mhadley157@gmail.... Re: HyCas
Hi. May I add a few points regarding the hycas board. Firstly, other builders of this board report seeing a gain of between 45 and 50db loosing around 4db when operated on 12.000 MHz (when changing the input L match only).

Because of the high impedences seen around the input stage it is absolutely essential that measurements are taken with the board in a fully screened enclosure and effective decoupling of supply and ancillary outputs.

I also strongly recommend the use of a post IF filter. In my case a four pole crystal filter with a 4.000 khz bw. It does make a big difference with plenty of output left to drive a diode mixer.

NB. If you are not aware, I supply a commercially produced fully updated pcb with through plated holes, screen printed component layout, solder mask and no board errors. I also have a couple of full kits still available.

Mike Hadley G4JXX mhadley157@gmail.com for details

Sent from my ASUS

-------- Original Message --------
15330 2019-06-04 11:30:02 jim_amos_n8cah Re: HyCas
My test results so far are mirroring what others are getting.  I see good AGC action, but overall gain is limited to at best 45dB.   I had it close to 45 dB by playing with the input inductor, and retuning.  But, I took two more turns off of the inductor, know it's a solid 40dB!  

I removed Q10 to disable the AGC Action, and to allow me to still use the on board pot, R32 to adjust the gain.  This works well and allows me to have a manually adjusted amplifier.   I set the gain to 40dB (Max after the input experiments still at this point).  I then started trying different values of components at different locations to see if I could effect the gain.     The gain distribution of my amplifier is currentonly +13, +9, +16 dB for the three stages.     Since the middle amplifier seemed to have the lowest gain, so I started to experimenting with different component values and parts substitutions.   So, I've swapped out the 2n3904, J310, input R (for 3.3K), output R ( for 3.3K) but no change in gain.

To my surprise so far, I have not been able to see any change in the basic 40dB of gain that I originally had.  This tells me that something else is limiting the gain so far.   

I agree with the idea that the HyCAS needs a shielded enclosure.  I plan on doing that, but my 40dB gain limit shouldn't be caused by a lack of enclosure.   

I am currently also building a simulation.   The basic circuit with manual gain, and standard models for the 2n3904 and J310 is getting +90 dB.  So, the extending this to include AGC as well, but this is not completed.

I'm going to experiment with shielding and a few other things to see if I can get the manual gain off of the +40dB.    I also want to look more closely into the gain that is being supplied by the differential amplifier.    I might also deadbug a stage to see what gain I get from it.

73's Jim N8CAH
15331 2019-06-06 11:34:06 jim_amos_n8cah Re: HyCas
I did a frequency sweep and what I saw was that at 1MHz I had 10 more dB of gain (50 dB total), and at 6-8 MHz I had about 5 more dB of gain (45 dB total).  So, there' clearly a frequency dependent element, the 40dB of gain that I was reporting was only at 9MHz.  But, this looks quite a bit different then then what is shown in the artical, I don't have a peak that is tuned by the input circuit.

By using a frequency sweep, I can see where I'm really just moving the corner of a LP filter.  So I have two things to do: 1) rewind my input inductor and recover the input peaking 2) wind some new higher Q 47uH or 100 uH chokes
15334 2019-06-08 05:20:08 jim_amos_n8cah Re: HyCas
I removed L2 and L3 and found that they were actually 470 uH inductors, instead of the 47uH in the design.   I don't think I had an easy way to measure these when I build the board, but now have one of the Chinese LC meters.   So, I've been using that tool to test my inductors that I wind!

I tried 11 turns on a FT-50-43, and 12 turns on a FT-37-43 with similar results: I'm know getting about 51.5 dB of gain with no AGC enabled, all inductors were measured at about 50uH.  I similarly rewound L1  for 7uH and know getting a good peak at 9 MHz.  

I did some simulations that showed that the added capacitance of the small larger value inductor lowered the gain of circuit similar to what I was seeing on the bench.   Larger values at R1 and R2 should increase the circuit gain, but since this is an AGC amplifier, I'm not sure what this does to the AGC Range.    There are also indications that the circuit is not stable at higher gains without a lot more shielding, so increasing gain further may not be useful here.  I may also change R1 and R2 back to 2200 ohms.

My next step is to re-enable AGC, and test gain vs. AGC voltage again.  I may also swap out L4 to a similar 47uH inductor, but I don't think that is critical. 

73's Jim N8CAH

15351 2019-06-11 07:24:52 jim_amos_n8cah Re: HyCas
I plotted Pin vs. Pout, Gain, and Vagc for gain settings of 52.5dB (max) and 40 dB plots can be found here:

I've provided both a PDF and the original spreadsheet I used to generate the plots.

The gain activation point changed with the gain setting.  So, for the lower gain setting I measured (40dB) AGC started limiting gain at -73 dBm.  For the 52.5 dB gain setting, AGC action started at -92.1 dBm.   

Looking at the plots that I generated, the gain, and AGC voltage plots both have a very distinct corner characteristic where AGC starts.  Also of interest, was that the gain (or attenuation) was predictable all the way up to +10 dBM input power.  

The amplifier is very stable in or out of an enclosure.  But, outside of an enclosure it is prone to picking up noise and other signals (the BFO signal for example).  I was seeing a lot of hash centered around 8 MHz when doing sweep plots and narrowed the source down to my oscilloscope.  If I had my scope leads draped over the amplifier it would pick up noise coming off of the grounds on my scope probe.    

So, after fixing my issue with the inductors, I was able to see 52.5 dB of gain.  I did see up to 55 dB of gain by  changing R1 and R2 to 3300 ohms.    The HyCAS design continues to be a great IF Amplifier design with very predictable behavior.   
15355 2019-06-11 08:07:47 bwbangerter Re: HyCas
Jim, thank you for uploading the data and plots for the measurements you made on your HyCas circuit. How did you adjust the open-loop gain (your 52.5 dB and 40 dB settings), with the IF gain control R32? I find that adjusting R6, the Threshold Set value (a 500 ohm trimpot in my circuit), sets the circuits output level when AGC is operative. I think you are using a fixed value resistance at R6? I am presently away from home, and unable to make measurements on my HyCas circuit for a week or so.

Ben, K0IKR
15357 2019-06-11 08:52:21 jim_amos_n8cah Re: HyCas
Hi Ben,

I've got a signal generator, and a spectrum analyser.    The signal generator is Marconi that allows me to set the output level in 0.1 dBm increments.  The Spectrum analyser provides the output power of the amplifier.   It's obviously pretty easy with that kind of test equipment.   So, the output power is measured directly in dBm for me.   I did get this equipment surplus - it works well and makes the measurement part much easier. 

What kind of test equipment do you have available?   

A signal generator (based off of one of the low cost chips such as a si570, si5351, or AD9850 works well.   If you couple this with a stepped attenuator (which can be fairly straight forward to build) you would actually provide similar capabilities to the Marconi for this measurement.   You can measure the output power with an oscilloscope by converting the output voltage to power if you terminate the output of the amplifier into 50 Ohms (a 50 ohm resistor).