Standard Operating Level for components in Audio Chain?

Discussion in 'Amplitude Modulation - AM Fans' started by W8KHK, May 25, 2018.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: MessiPaoloni-1
ad: L-MFJ
ad: Subscribe
ad: Left-3
  1. W8KHK

    W8KHK Ham Member QRZ Page

    I have been out of town all month. Here is the schematic from which the board was created by EasyEDA. I know everyone made lots of suggestions and provided help while I was testing the sections on the proto boards. I truly appreciate all the input.

    Since this was my first attempt at creating a PCA, I went a bit overkill on options, such as extensive bypassing and decoupling, jumpers to section the board in case all the functions did not play well. This gives me many options to include or exclude some of the mods that did or did not play well while testing each section on the bench. For receiver use, minimal external controls are needed, but for the transmit function I wished to be able to enable or disable each of the filter sections, depending on their performance and band conditions. So headers were installed so that dupont wires could be interfaced to switches and pots. If all functions are desired, either berg jumpers, or just a wire jumper, may be used to interconnect the various sections.

    The LM833 did not play as well as expected, but the OP275 worked flawlessly as a differential line driver. At test time, I had only SMD chips, so a carrier was inserted in the respective socket. I have since received a package of OP275 devices with legs to go in the remaining PCAs.

    After testing, I determined all functions played well except the LM386 amplifiers, and for those the jury is still out. I did intend to leave out two of the electrolytic capacitors. I attempted to use plus and minus 5 volts for the LM386 instead of a single 12 volt supply. I may need a cap between the pot and the input pin. If it works out, the output coupling capacitor location may be bypassed with a jumper.

    Attached Files:

    Last edited: Aug 18, 2018
  2. W8KHK

    W8KHK Ham Member QRZ Page

    Here is a view of the populated board, as well as a view of the board with the companion switch mode power supply. More will be forthcoming as testing is completed and the board is packaged with controls. 20180817_222247.jpg 20180817_222306.jpg
  3. K7JEM

    K7JEM Ham Member QRZ Page

    I think you will need that capacitor between the pot and the input pin. Pin 2 on each amp will probably need to be referenced to -5v instead of ground in this application, since the 386 is designed to run from a single voltage source. You may not need the output capacitors when that is done, the voltage should be near zero. But it would be good to verify that.

    But the pictures look good, very professional.
  4. W8KHK

    W8KHK Ham Member QRZ Page

    I did thorough testing of the LM386 stage on the proto board today, both with a single 12 volt supply, and split plus and minus 5 volt supplies. Apparently there was an error in my original test setup, because my prototype schematics indicated both pin 2 (negative input) and pin 3 (positive input) could be referenced to ground when the split supply was used. Testing has revealed that is not the case.

    I found it necessary to reference both pin 2, and the bottom of the pot, to the negative rail, ie pin 4. If either one is left floating, or not referenced to the negative rail, there is no output. A coupling capacitor between the pot and pin 3 also causes failure.

    When using the split supply and referencing pin 2 and the pot to pin 4, the output is normal, and the DC offset on pin 5 is less than 2 millivolts, assuming the supplies are balanced.

    On the finished PCBs, it will be necessary to lift the bottom of the pot and pin 2, and reconnect both to pin 4, or it will be necessary to disconnect pins 4 and 6 from the negative and positive supplies, and connect 4 to ground and 6 to plus 12 volts. This latter solution is relatively easy to implement, then the output capacitor may be installed. If it is possible to implement the former solution with minimal board hacking, then the output capacitor may be jumpered.

    Thanks again to all who assisted. I feel fortunate that there was only one correctable fatal flaw in the design, as it was a rather ambitious project for a first attempt at creating a PC board using software. I did some PC boards back in the 70s for audio amplifier projects using black tape and transparent mylar at 5x actual size, then reduced photographically. We have come a long way with the software approach. The amazing thing to me is that the entire circuit design and layout was completed with the web browser. No local software was needed! If anyone here is interested in taking the plunge to design their own PC boards, I would highly recommend they consider using EasyEDA. The software side is free, board creation is extremely inexpensive, and turnaround is amazingly fast. Bottom line, I know I will not be building any more relatively complex circuits with wire-wrap wire and tweezers!
    Last edited: Aug 22, 2018
  5. W8KHK

    W8KHK Ham Member QRZ Page

    The receive version of the audio filter is completed and tested. The transmit version is identical electrically, but will be packaged differently. It will include connections for both single-ended and balanced differential input and output. It will have a calibrated VU meter, headphone/speaker output, and ability to enable or disable any or all filter sections.

    The receive version is packaged in a smaller enclosure, with only single-ended plated RCA connectors for input and output. Switches allow the entire filter to be turned off and disabled, as well as enabling either the SCAF, the anti-aliasing low-pass filters, or both. SCAF cutoff frequency is adjusted via a front-panel 2K pot.
  6. W8KHK

    W8KHK Ham Member QRZ Page

    Here are a few pictures of the innards. The main filter board was created online with EasyEDA, and produced by JCLPCB in china. The power supply ( $9 kit) is a switcher from It will function with a DC input from 5 to 25 volts DC, and provides + 3.3, - 5.0, + 5.0. - 12.0, and + 12.0 volts via linear regulators.

    The wiring looked rather neat and organized until I added the leads to the power supply mounted in the cover. But when the screws are holding it together, no one will see the cob webs.

    When the transmit version is packaged, additional photos will be posted.



    Last edited: Sep 13, 2018
  7. N2DTS

    N2DTS Ham Member QRZ Page

    Looks very nice!

    I built a logikit scaf-1 with the AB2EZ mods and it works well but its unbalanced and when open to maximum does some odd stuff.
    W8KHK likes this.
  8. W8KHK

    W8KHK Ham Member QRZ Page

    Hi Brett,

    It was Stu's write-up that piqued my interest in making one. So I started hand-wiring the Idiom Press schematic on perf-board, including Stu's mods. I was also not happy with some of the artifacts, so I have been working on this project on and off since May. I wanted something that would clean up the signal in my audio chain before I put the plate-modulated PP 250-TH rig on the air. I especially wanted to guarantee that I could set it to avoid unnecessary signals above 3.5 and 4.0 KHz (when band conditions warrant a narrower signal). I have no other brick-wall filter.

    I spent quite a bit of time with the HP sine and square wave generators and the Tek scope, and read a great deal on op-amps, especially the writings of Walter Jung. I also leveraged some of the filter design Steve, WA1QIX put in his PWM modulator board. Another problem area I found with my implementation of the Idiom design was instability in the NE555 clock, and I addressed that as well. I am finally happy with the results, so I will use one for receive, and another one dedicated to transmit functions. My friend and retired HP colleague Rick (W4BYT) worked with me from start to finish on this project. We continued to troubleshoot and refine the circuit until we were both happy with feeding it into a modulator!

    I have extra boards and lots of parts, so if you are interested in making one, just let me know. My first PCB attempt with software did not come out perfect - I needed to cut one trace and add one jumper, but other than that all is well. The only (relatively) expensive part is the Maxim MAX295 device. The case, under $5.00 from M P Jones was a perfect fit. I found that long after the boards were produced.

    73, Rick
  9. N2DTS

    N2DTS Ham Member QRZ Page

    I also like the idea of higher power transmitters being able to limit the bandwidth when its called for.
    I would be interested in a TX version board(s).

    Looks like it was a fun project.

    I am not sure what happened when I opened up the scaf filter all the way, I had some sort of gain/feedback start at the high frequency limit.
    And sometimes I would bang the thing with my hand and it would go away.
    I had RF filtering and went over the solder joints and connections carefully, but it always did it (sometimes).
    I really stopped using it because its unbalanced.

    push pull 250 TH rf deck!

    I heard Borgi saying something about some new project last night, likely big tubes involved...
    W8KHK likes this.
  10. W8KHK

    W8KHK Ham Member QRZ Page

    Brett, I sent an email your way...

Share This Page