Topics

uBitx has a bit too crispy / semi-distorted audio even on low signal levels #ubitx #ubitx-help

Mike
 

I have completed the uBitx and calibrated both frequency and BFO. The sensitivity appears OK. However audio has a strange crispy / semi-distorted sound to it.

I read the post from W3JDR https://w3jdr.wordpress.com/2018/04/11/ubitx-dynamic-range-improvement/#comment-1507  about biasing Q70 more strongly, however that only reduced the gain somewhat but didnt seem to help the crispy / semi distorted sound.

I have tried several headsets and a speaker. Same issue.

Anyone have any hints to what may be done to achieve better audio and get rid of that distortion effect?

KE6BB
 

Are you sure you have the BFO set and fine tuned?   Check out VK3YE's YouTube video to see how he fine tuned his BFO setting:

https://youtu.be/sXJmAhpAjeI

He is listening to his transmissions, but if my memory hasn't failed me, the setting affects both TX and RX.

Hope that helps.
--
Mark,
KE6BB

Ralph Mowery
 


My receive and transmitted audio did not have much highs or lows in it.  The filter was only about 1700 Hz wide.  I changed the 5 capacitors from the crystals that are used for the filter from 100 pf to 82 pf ( think that is the standard value) and it widened out the audio range .  Probably should parallel some 5 or 6 pf capacitors across them to get about 85 to 88 pf as I did not have anything between a low 80 and the 100 pf.

That effects both the transmit and receive response, so you may want to make sure the transmit audio sounds about the same.

de ku4pt


On Thu, Jun 14, 2018 at 1:18 PM, Mike <rezycle.bin@...> wrote:

I have completed the uBitx and calibrated both frequency and BFO. The sensitivity appears OK. However audio has a strange crispy / semi-distorted sound to it.

I read the post from W3JDR https://w3jdr.wordpress.com/2018/04/11/ubitx-dynamic-range-improvement/#comment-1507  about biasing Q70 more strongly, however that only reduced the gain somewhat but didnt seem to help the crispy / semi distorted sound.

I have tried several headsets and a speaker. Same issue.

Anyone have any hints to what may be done to achieve better audio and get rid of that distortion effect?

_

Jerry Gaffke
 

If the audio is loud enough and has sufficient bass and treble and an SSB transmission is intelligble,
then this is not likely a BFO issue.  Could be a fault most anywhere between the antenna port
and headphone jack.   Though I would expect this problem to be somewhere from the demodulator
on through the audio amps.  

I would probably first try injecting a low level audio signal into the bottom of C50.
Could use audio from an mp3 player or broadcast radio headphone jack, but reduce the 
power level by a factor of a million or so (60 dB) first to avoid over driving Q70.
That means a voltage ratio of sqrt(1000000) = 1000  so from headphone jack to a 1 meg resistor
to a 1k resistor to ground, add a wire from the top of the 1k resistor to the bottom
of C50.  You also need a ground connection.  If that sounds good, then trouble is further back in the radio.
I'd look very hard at the demodulator if it isn't in the audio stages.
Trouble in the RF sections of the radio might mean birdies or weak signals, not so likely to 
create distorted audio.

Not sure anybody here knows what objectionably "crispy" audio sounds like,
you might give a link to an audio recording.

If there's another uBitx or Bitx40 in the neighborhood, you could patch in a signal from that other
radio into the your rig.  For example, inject a signal into c30 on your failing radio from the corresponding
point of that other radio, if the receiver now works well that means the fault is closer to your antenna jack.
Ideally remove c30 from both rigs, then have a 0.1uF cap in series with the
signal wire from one rig to the other to keep impedances about where they should be.  That wire is 
ideally a piece of 50 ohm coax, ground both ends of the coax braid. 

Good luck!

Jerry


On Thu, Jun 14, 2018 at 10:18 am, Mike wrote:
I have completed the uBitx and calibrated both frequency and BFO. The sensitivity appears OK. However audio has a strange crispy / semi-distorted sound to it.

I read the post from W3JDR https://w3jdr.wordpress.com/2018/04/11/ubitx-dynamic-range-improvement/#comment-1507  about biasing Q70 more strongly, however that only reduced the gain somewhat but didnt seem to help the crispy / semi distorted sound.

I have tried several headsets and a speaker. Same issue.

Anyone have any hints to what may be done to achieve better audio and get rid of that distortion effect?

Mike
 

Hi I actually have two uBITx radios. One that I have tuned the BFO on and another factory stock. I did swap in the other uBITx into the same cable harness: the "crispy / semidistorted" audio is exactly the same on both of them. Its kind of "cracking up" on the peaks. If i listen to a carrier, I hear the same effect. The tone is not clean.

Mike
 

Here is an example of how it sounds like. SSB signals kind of sounds like someone runs a strong RF clipper on them. If you wait until the last part of the clip you can also hear CW signals. Those signals are not too strong and should not distort.

https://groups.io/g/BITX20/files/180615_1218.MP3

Mike
 

ajparent1/KB1GMX
 

There is distortion, sounds like clipping in the receiver due to overload
by a strong signal.

Also check to see if the Q70 mod is needed, it is listed in a different thread.

The last items is maybe the speaker itself is bad and distorting.

Allison

Mike
 

Hi!

There is also overload on weak signals
Tried Q70 mod: that didnt help, still distortion
Tryed several speakers and several headphones: the same
Tried two different ubitx units: no change, same distortion
I received the units around 15june2018. They were sent in June 2018. Could this be a bad batch?
Anyone else have experienced this? Any more hints?

ajparent1/KB1GMX
 

Receiver overload.  There is no AGC in these.  TO my ear the CW sounded a bit hard.
Try tuning in a really weak signal.

It could be the audio amp.... V3 or V4?


Allison

GM4CID
 

Are you comparing  the uBITX signal quality with that from another receiver?
I could be wrong but S57DX sounds like a typical highly processed signal.

WA9PWR
 

i was thinking the same thing.  Should try other signals, NOT during a contest.

Mike
 

Disregard S57DX and listen later in the recording. Pls listen to the CW signal later in the recording. This is a weak signal. Even the very weak signals barely detectable sounds distorted.

Mike
 

Its V4

Mike
 

I have used an oscilloscope and a signal generator to more tests regarding the audio distortion.

I have followed the signal thru the audio chain after the last mixer when appying a known clean and weak signal on the input so no front end or mixer overload ocurs. Everything looks clean on the oscilloscope until the output of final two transistors. The signal on the output of Q71 looks fine. On the output of Q72/ Q73 the signal is distorted around the zero crossing. https://groups.io/g/BITX20/photo/59179/3?p=Name,,,20,1,0,0

I wonder, is the biasing actually correctly done?
Should it have been like this instead? https://groups.io/g/BITX20/photo/59179/0?p=Name,,,20,1,0,0


Here is the distorted signal out of the amp (blue trace, note different Y axis scale). The red trace is the signal earlier in the audio chain. https://groups.io/g/BITX20/photo/59179/1?p=Name,,,20,1,0,0

Pls post your comments.

Mike
 

Some more testing results:
I attached an external high quality Hi-Fi type NAD linear audio amplifier via an electrolytic capacitor (for DC blocking). On that NAD amp. the audio is no longer distorted. I have a feeling that the final amp of uBitx V4 has a flaw in it.....

As far as I know, this is the most common way of biasing a push pull amplifier to avoid crossover distortion: https://www.electronics-tutorials.ws/amplifier/amp110.gif
To me it looks like there is a problem in the final amp ...

Rahul Srivastava
 

Q70 appears to be the bottleneck with its extremely high gain try reducing R51 to 1k5 and R50 100k-150k. See  what works best.

Rahul VU3WJM

 

Mike,

Could you do a voltage test around the audio amp and report ? Volume = 0.

Q71/72/73 CBE

Raj


At 14/06/2018, you wrote:

I have completed the uBitx and calibrated both frequency and BFO. The sensitivity appears OK. However audio has a strange crispy / semi-distorted sound to it.

I read the post from W3JDR https://w3jdr.wordpress.com/2018/04/11/ubitx-dynamic-range-improvement/#comment-1507   about biasing Q70 more strongly, however that only reduced the gain somewhat but didnt seem to help the crispy / semi distorted sound.

I have tried several headsets and a speaker. Same issue.

Anyone have any hints to what may be done to achieve better audio and get rid of that distortion effect?

Mike
 

I now replaced the output amp with a MCP6284 that I had laying in the shack. That works.  No distortion. I tapped the signal from the collector of Q71 via an electrolytic cap. This strengthens my suspicion that there is something going on with the output amp.

One thing I also noticed now is that there is a clicking sound whenever I change frequency with the VFO .... Is this known? I wonder if it is due to the clock generator? Or crosstalk ...

I will also do as you suggest : measure some voltages around Q71, Q72, Q73 on BCE.

Mike
 

Raj, VU2ZAP: here are the voltages you asked for.