Topics

HI-PER-MITE parts question


Nick-WA5BDU
 

Thanks for the responses.  I did finally get it all together and into a box.  I think it'll be handy on things with no CW selectivity like the Argonaut 505 and my various homebrew projects.  I was concerned that I didn't want to foul up the frequency response while trying to adjust the gain.  I could see that R12 and R11 are just a voltage divider for gain control (reduction) into the LM386, but it looked like R1 also functions as part of the frequency determining network of the first stage.

Finally did an LTSpice model of the circuit to make sure I didn't make a mistake before soldering things together.  Interesting -- you can see the staggering described in the article.

I made some measurements of voltages needed to produce low, medium and high audio on typical earbuds and they're remarkably small.  A faint sound was 7 mV rms, 3.1 uW.  Moderate was 29 mV rms, 50 uW, louder was 57 mV rms, 0.36 mW.  This was into a 16 ohm earbud.

I substituted a 100k pot for R11 and R12 as suggested in the manual. 

Since line levels can be a couple of volts, I had some concern about overdriving the stages in some instances, so I added an 11:1 divider (10k series, 1k shunt) to the front.  There's still plenty of gain for sure even after that ~20 dB reduction.

72-

Nick, WA5BDU

n 4/22/2013 8:28 PM, Dennis Gaskill wrote:

 

Nick;
I think I remember what was going on ...........
 
If you look at the schematic, the resistors  R1, R2, R11, R12 are set up for 30 db gain. Refer to the table for setting gain on page 2 in the manual.
When they created the Center Frequency Table, I think they were looking at the schematic........
The only difference between 0 db and 30 db is that you swap R1 and R3...........
R3 and R11 and R12 are not in the Center Frequency Table, therefore R1 is all we have to worry about for gain.
One more factoid; the different center freqs are found by scaling the resistors by the same ratio as the freq ratio.
Example: If we are moving the center freq from 700 to 640 then the ratio is 700/640 = 1.094 
So if R1 was 33K at 700 Hz the R1  at 640 Hz would be 1.094 * 33K = 36 K
And at long last if R1 is 1meg * 1.094 = 1.094 meg for R1.............
I would just leave it at 1M .....it won't make that much difference, unless you are going for 500 Hz.
Sorry for the long winded explanation...............and sorry this took so long to reply.
 
For My money I would leave it at 700 Hz. Will 70 hertz make enough difference to justify scrounging up the new resistors?
I am just Lazy
Dennis KC0IFQ
 
Sent: Friday, April 19, 2013 6:39 PM
Subject: [4sqrp] HY-PER-MITE parts question

 


I picked up a HY-PER-MITE filter kit at OzarkCon and am getting ready to
put it together. Checking out the configuration options first.

The instructions for the center frequency say to use R1 = 36k for the
640 Hz option (versus 33k for the 700 Hz standard center).

The assembly instructions for setting gain say that for a gain of 0 dB,
R1 = 1M.

How do I reconcile the two for 640 Hz and 0 dB? Could one of these be
referring to something other than R1?

72-

Nick, WA5BDU










Dennis Gaskill <gaskilld@...>
 

Nick ;
Did you give a talk on LTSpice at OzarkCon a couple of years ago?
Anyway, old buddy, old pal would you like to share what you have modeled of the HY-PER-MITE.
Just might be purty handy! You might want to upload it to the Yahoo group. 
I am glad you have tamed the HPM beast....... Good work!
 
Dennis KC0IFQ
 

Sent: Thursday, April 25, 2013 2:39 PM
Subject: Re: [4sqrp] HI-PER-MITE parts question

Thanks for the responses.  I did finally get it all together and into a box.  I think it'll be handy on things with no CW selectivity like the Argonaut 505 and my various homebrew projects.  I was concerned that I didn't want to foul up the frequency response while trying to adjust the gain.  I could see that R12 and R11 are just a voltage divider for gain control (reduction) into the LM386, but it looked like R1 also functions as part of the frequency determining network of the first stage.

Finally did an LTSpice model of the circuit to make sure I didn't make a mistake before soldering things together.  Interesting -- you can see the staggering described in the article.

I made some measurements of voltages needed to produce low, medium and high audio on typical earbuds and they're remarkably small.  A faint sound was 7 mV rms, 3.1 uW.  Moderate was 29 mV rms, 50 uW, louder was 57 mV rms, 0.36 mW.  This was into a 16 ohm earbud.

I substituted a 100k pot for R11 and R12 as suggested in the manual. 

Since line levels can be a couple of volts, I had some concern about overdriving the stages in some instances, so I added an 11:1 divider (10k series, 1k shunt) to the front.  There's still plenty of gain for sure even after that ~20 dB reduction.

72-

Nick, WA5BDU

n 4/22/2013 8:28 PM, Dennis Gaskill wrote:
 

Nick;
I think I remember what was going on ...........
 
If you look at the schematic, the resistors  R1, R2, R11, R12 are set up for 30 db gain. Refer to the table for setting gain on page 2 in the manual.
When they created the Center Frequency Table, I think they were looking at the schematic........
The only difference between 0 db and 30 db is that you swap R1 and R3...........
R3 and R11 and R12 are not in the Center Frequency Table, therefore R1 is all we have to worry about for gain.
One more factoid; the different center freqs are found by scaling the resistors by the same ratio as the freq ratio.
Example: If we are moving the center freq from 700 to 640 then the ratio is 700/640 = 1.094 
So if R1 was 33K at 700 Hz the R1  at 640 Hz would be 1.094 * 33K = 36 K
And at long last if R1 is 1meg * 1.094 = 1.094 meg for R1.............
I would just leave it at 1M .....it won't make that much difference, unless you are going for 500 Hz.
Sorry for the long winded explanation...............and sorry this took so long to reply.
 
For My money I would leave it at 700 Hz. Will 70 hertz make enough difference to justify scrounging up the new resistors?
I am just Lazy
Dennis KC0IFQ
 
Sent: Friday, April 19, 2013 6:39 PM
Subject: [4sqrp] HY-PER-MITE parts question

 


I picked up a HY-PER-MITE filter kit at OzarkCon and am getting ready to
put it together. Checking out the configuration options first.

The instructions for the center frequency say to use R1 = 36k for the
640 Hz option (versus 33k for the 700 Hz standard center).

The assembly instructions for setting gain say that for a gain of 0 dB,
R1 = 1M.

How do I reconcile the two for 640 Hz and 0 dB? Could one of these be
referring to something other than R1?

72-

Nick, WA5BDU












Nick-WA5BDU
 

I sure did.  Now two years later I've forgotten a lot of what I learned preparing for that talk. ;^)

Actually, I still use LTspice a lot but it does have its complexities, as all SPICE programs do.

Anyway, I finally beat my analysis of the HI-PER-MITE into shape such that I'm not too ashamed to show it and have posted it on the 4SQRP Yahoo Group's site.  I created a new folder under "files" for the LTSpice files.  Take an look and if there are any major problems, let me know.

73-

Nick, WA5BDU

On 4/25/2013 4:08 PM, Dennis Gaskill wrote:
 

Nick ;
Did you give a talk on LTSpice at OzarkCon a couple of years ago?
Anyway, old buddy, old pal would you like to share what you have modeled of the HY-PER-MITE.
Just might be purty handy! You might want to upload it to the Yahoo group. 
I am glad you have tamed the HPM beast....... Good work!
 
Dennis KC0IFQ
 



Wesley Matthews
 

Nick-

could you share how you wired one pot to substitute for 2 resisters please? I’m at that stage right now in a Century 21 filter upgrade. 


Thanks
--
73

Wes
W3KW

Happy Hamming.


nm0s_qrp
 

Check out the schematic for the SSB-Mite.  It shows how to do it explicitly, and you can just bring that mod directly over to the HiPerMite.