EQ before or after the Processor

Conny

Reading thru some of the conversations I got mixed information about how to adjust my modulation. Some indicates that it's important that you set EQ not to over drive the processor. 

Then I found this in the manual:

"Adjust the TX Equalizer to compensate for the microphone and operating conditions. The default setting is a good starting point for most microphones. The TX EQ is located after the processor in the signal chain, so it should typically be adjusted prior to engaging the processor. Minor adjustments can be done after engaging the processor, but care is needed to avoid excessive peaks."

If this is correct, EQ is applied after processing, I'm not sure how adjusting the equalizer would have any impact on the processor output. I can see how adjusting the EQ could mitigate some of the artifacts that the processor might have added to the voice.

73
N5HC
/Conny

Steve

Since the processor in this case is a compressor EQ should be done first.  I operate AM most of the time.  I use a Symetrix 528E that I use in front of the Flex.  It comes with the compressor ahead of the EQ,,WRONG..it is able to be "reordered" and so now the compressor comes after the EQ. 

Conny

I understand with an outboard EQ you will always be before the internal processor, if used. But as I understood it by reading the manual: The TX EQ is located after the processor in the signal chain, and that would be the internal signal chain. Meaning changing the internal EQ settings can't really "over drive" the internal processor. Did I miss something here?

Steve

I never really thought about the order of things within the Flex.  I do not use the EQ within the Flex.  I process outside and dump the audio in.  Sorry if i am not much help.  If the EQ is after the processor it is wrong...

Bill -VA3WTB

A few of us have spent a lot of time studing this.

The EQ settings have a direct effect on the processor. For this reason some of us do not adjust the sliders over 0 and run them below 0. Or maybe +1 or +2 at the most.

A couple of us have been talking to the Flex engineering about what we now know.

One problem as we see it is there is no meter reading reflecting what gain the EQ is having on the processor. Lets say you have the processor on and the mic gain set so you see -5 on the voice peaks. If you set the EQ sliders to high the voice peaks can go into the red,, but the peak meter is not showing this. The Meter is not showing the EQ gain added to the chain. Distortion creeps in without us knowing.

In PSDR in the EQ section there was, Is a EQ gain control. If you set the meter to the EQ reading it will tell you what the EQ gain is set at on the meter scale. This way we can then see if the EQ is peaking at 0. Then after everything is set we want the ALC peaking around 0.  The problem for SSDR, is we can't see the gain added by the EQ. We are asking Flex to have the peak meter reflect both the mic gain and the EQ gain together showing on the meter for better setting.

Conny

If EQ have a direct effect on the processor then the manual is misleading as it says the TX EQ is after the processor. Unless there are some sort of feedback/control after the EQ back into the processor. 
I'm new to the 6600m and I'm trying to understand as much as possible what's going on under the "hood".

Bill -VA3WTB

Think of it this way, The EQ is not just EQing Freq,, each slider works as a gain control. Each slider, increases gain to the mic chain.

I can't explain how, I just know it does this. Interesting, I have heard a few Flex radios starting to distort, then come to find out were their EQ sliders are set, many are around +5 +7 and so on driving the audio to hard.

Jay Nation

Has everyone wondering about the EQ stage, read this. It's great advice.
Cutting before Boosting, will work a lot better.
https://helpdesk.flexradio.com/hc/en-us/articles/203853305-Rules-for-EQing-Voice-for-Optimal-Phone-Operation

Unless somebody links to it, some may never know it's there.

#FlexRadio IRC Chat

   73, Jay - NO5J

Mike-VA3MW

Jay nailed it... cut before boosting.  

K0FLY

 Jay: 

 I'm like you, but I cut to much, all of the sliders were below the zero line.  The locals on 75 told me that my KWM-2/ S Lines and Elecraft had more talk power than the new Flex. 

After a few days I tried one of the default Flex mic profiles, up came the Flex talk power,  the Flex is now as  good or better than Collins.  What I finally figured  out was that the EQ setting levels affect the transmitter drive.  I increased the 250 to 4 kHz sliders so that they are now in the 4 to 5 range with the 2 kHz at 7.  Keep your  EQ profile but increase each slider by the same amount until you get your talk power up

With -10 to -5 on the mic input meter, I now get 3 to 5 dB on the compression meter (normal processor setting).  The locals like this setup.   I don't like a lot of compression but I think that you need some.  I remember the guys on Navassa telling the Europeans to shut off their processors that they were unintelligible.

Gayle K0FLY 

Conny

Definitely understand the preference of cutting over boosting.

But it is still unclear to me if the internal EQ is applied before or after the processor.  Especially when the internal processor is a CESSB type of processing.
 
N5HC
/Conny

Mike-VA3MW

I thought you quoted it at the beginning.  

CESSB is done in the RF station, not the AF stage, so well after the audio processing.  It can only be done at the RF level.

Conny

The manual(s) ambiguity in this matter and lack of even the basic signal flow diagram make me feel like this:

   I know if I do this, this will happen, but I don't really understand why, it's like magic.

Don't like to operate that way. 

Hence the detailed question about what the manual refer too in regards to what EQ before and after the processor means. And now, perhaps the question should be expanded to included what the processor (in the manual) refers to too. Is it the AF processing or the CESSB processing or both?

Still new to the Flex concept.

Bill -VA3WTB

Jay, that could possibly work, how ever you have no way of knowing if you are starting to distort, as you increase the sliders, the gain is not shown on the meter.

A rule of thunb is a setting at 0 on EQ will give good talk power, if you need a little more moving to +2 or +3 is lots.
I run most my sliders down around -3 to -5. The 63 is set to 0 and 2k is at +1 4k Is at +3 as an example.

In order to get max talk power I increase the mic gain to -5 peaks.
With cuting, keeping the sliders lower, I can now use DX and DX+ without a hint of distortion.

Steve

Very interesting...perhaps what needs to come with these radios is a block diagram and a complete section on "theory of operation", you know, like the old days????   I am sure Mike will suggest it to engineering.  Lots of speculation here and plenty of articles available on the net that more than adequately address EQ'ing and audio processing.  The missing part is how does a particular radio handle these tasks.  I am an AM operator and have adjusted my equipment (including more than one Flex radio) to have not just good but superb audio and I would not have it any other way..73's all

Jay Nation

Bill
Your method, you described above is exactly what I do. You say you are cutting on some frequencies and boosting on others, so you seem to also cut a little more often than boost. I tend to rely on on-air reports or recordings when I can get them, to know when I've boosted something into distortion. I generally expect to see something between -5 dBm and 0 dBm on peaks, and always stay in the yellow on the meter.

#FlexRadio IRC Chat

   73, Jay - NO5J

Mike-VA3MW

Don't forget, you can use the FDX feature to listen to yourself in another Slice.  This is the best way to set up your audio to your liking, not someone else's.