EZTube UA 610 EQ

[ruffrecords]

In my continuing search for solutions to the Baxandall quandary, ( http://groupdiy.com/index.php?topic=56238.0 ) I re-examined the UA610. Its switched EQ is principally in the NFB loop of the output amplifier. It occurred to me it might be possible to do something similar around the NFB of the output amp of the EZTubeMixer channel amp. It turns out that the basic NFB topologies are identical. Only component values are different. I did a few simple sims and it seems to work. I notced that the original +- 6dB EQ range has been extended in the ne UA2-610 to +- 9dB and both bass and treble ranges have three selectable frequencies. This complicates things a bit but I think it should be feasible. If there is some interest I will take it further.

Cheers

Ian

 

[Holger]

Ian, will it fit in a 3 U module?

 

[okgb]

On my  U.A. 6176 I find the eq occasionally , slightly useful , two shelves really

 

[ruairioflaherty]

On my  U.A. 6176 I find the eq occasionally , slightly useful , two shelves really

I'd be surprised if the modern versions used anything like the same topology, have you ever had a look inside?

 

[ruffrecords]

Ian, will it fit in a 3 U module?

Yes and no!. As the EQ is connected to the feedback network of the second amp I would be reluctant for it to be fed to a separate module. So I had envisaged it being confined to the mic pre module. However, the attached picture shows the insides of the UA 2-610. At just to the left  of centre at the bottom are two front panael mounted PCBs that hold he EQ components. One has a only a couple of toggle switches and the other contains two rotary switches and the EQ components. The toggle PCB appears to be connected to the EQ PCB by a couple of ribbon cables. The EQ board is then connected to the 610 mic pre circuit pair a pair of screened cables (blue). These must be at least eqaul to the distance between two 3U modules in an EZTube mixer so it looks as though it should be possible. If I designed it as a 3U  PCB the it could be used either in a separate module or inside a mic pre module.

Cheers

Ian

 

[ruffrecords]

On my  U.A. 6176 I find the eq occasionally , slightly useful , two shelves really

I'd be surprised if the modern versions used anything like the same topology, have you ever had a look inside?

From looking at pictures of the insides of the 2-610 it looks like the topology is very similar to the original.

Cheers

Ian

 

[Holger]

Ian, will it fit in a 3 U module?

Yes and no!. As the EQ is connected to the feedback network of the second amp I would be reluctant for it to be fed to a separate module. So I had envisaged it being confined to the mic pre module. However, the attached picture shows the insides of the UA 2-610. At just to the left  of centre at the bottom are two front panael mounted PCBs that hold he EQ components. One has a only a couple of toggle switches and the other contains two rotary switches and the EQ components. The toggle PCB appears to be connected to the EQ PCB by a couple of ribbon cables. The EQ board is then connected to the 610 mic pre circuit pair a pair of screened cables (blue). These must be at least eqaul to the distance between two 3U modules in an EZTube mixer so it looks as though it should be possible. If I designed it as a 3U  PCB the it could be used either in a separate module or inside a mic pre module.

Cheers

Ian

Thanks Ian

 

[ruffrecords]

Ian, will it fit in a 3 U module?

Yes and no!. As the EQ is connected to the feedback network of the second amp I would be reluctant for it to be fed to a separate module. So I had envisaged it being confined to the mic pre module. However, the attached picture shows the insides of the UA 2-610. At just to the left  of centre at the bottom are two front panael mounted PCBs that hold he EQ components. One has a only a couple of toggle switches and the other contains two rotary switches and the EQ components. The toggle PCB appears to be connected to the EQ PCB by a couple of ribbon cables. The EQ board is then connected to the 610 mic pre circuit pair a pair of screened cables (blue). These must be at least eqaul to the distance between two 3U modules in an EZTube mixer so it looks as though it should be possible. If I designed it as a 3U  PCB the it could be used either in a separate module or inside a mic pre module.

Cheers

Ian

Thanks Ian

My Pleasure. and by the way, I forgot to thank you for the couple of power distribution PCBs you sent me.

Cheers

Ian

 

[ruffrecords]

I have been looking into this in a bit more depth. One thing I noticed is that the cut/boost switches they use have seven positions - 3 boost, 3 cut and a centre off. Seven is not a convenient number if you want to use cheap switches like Lorlin. It looks at the moment as though the top cut and boost can be done with a single pole switch so 7 positions is just a stopped down 12 way switch. However it looks like the bass boost/cut needs two poles which is a bit of  pain. So one alternative is to have two steps for bass cut, for example and three for boost so we could use a regular 2 pole 6 way switch. Opinions??

Also, I wonder if anyone has a 2-610 from which they can measure the EQ curves for me? It is OK for the UA to put boost/cut frequencies on the front panel but, as there is no generally accepted method of defining shelving EQ frequencies, there is some doubt as to the response they will produce.

Cheers


Ian

 

[ruffrecords]

I don't know how I missed this but the UA 2-610 EQ has 1.5dB steps so there is a total of 11 positions so it looks like we'll need at least one 2 pole 11 way switch so its probably going to be Grayhills again.

Sims are looking promising. I'll post a schematic when I get close.

Cheers

Ian

 

[ruffrecords]

First round sims all done and it looks promising. All the basic boost/cut components are identified, just need to set frequency dependent parts for correct turnovers.

Cheers

Ian

 

[Joechris]

Great work, can this eq arragement be used with variable gain, like a mic pre, or does the gain need to be fixed?

 

[ruffrecords]

Great work, can this eq arragement be used with variable gain, like a mic pre, or does the gain need to be fixed?

At the moment, it is intended as part of the EZTubeMixer project. Normally, this uses passive EQ and the second amplifier on the mic pre PCB is used as a gain make up to compensate for the losses in the passive EQ. The idea with this EQ is that is fits in the negative feedback loop of the second amplifier instead of a passive EQ in front to of it. The current design has a gain of close to 16dB when all the controls are flat. This is intended to provide 10dB in hand on the preceding channel fader plus 6dB to allow for the 6dB loss in the direct out transformer (VTB2291 wired 2K4:600). The mid band gain is set by R14 (15K) and R19 (2.7K). Unfortunately, changing R15 upsets the dc conditions of the tubes and changing R19 will alter the turnover points and amount of cut/boost of some of the EQ settings so making the gain variable is not really an option.

Cheers

Ian

 

[ruffrecords]

A kind group member has provided me with wav file sweeps of the response of his 2-610. I thought it would be easy to turn these into frequency response plots using LTspice but this turns out not to be the case. I can import the files easily enough but they can only be used as a stimulus.

So, does anyone know how to convert some wav files  of frequency response sweeps into frequency response plots so I can work out the turnover frequencies for this project? I think some sort of FFT is in order or perhaps some spectrum ananysis. All suggestions welcome.

Cheers

Ian

 

[ruairioflaherty]

Hi Ian,

Are you on Mac or PC?

Can you email me the sweeps or a link to them?  I can try to import to Fuzzmeasure.

Cheers,
Ruairi

 

[culteousness1]

So, does anyone know how to convert some wav files  of frequency response sweeps into frequency response plots so I can work out the turnover frequencies for this project? I think some sort of FFT is in order or perhaps some spectrum ananysis. All suggestions welcome.

This will take about 5 lines of code in MATLAB.
SCILAB will do too and it's open source.

Cheers,
Carsten

 

[ruffrecords]

So, does anyone know how to convert some wav files  of frequency response sweeps into frequency response plots so I can work out the turnover frequencies for this project? I think some sort of FFT is in order or perhaps some spectrum ananysis. All suggestions welcome.

This will take about 5 lines of code in MATLAB.
SCILAB will do too and it's open source.

Cheers,
Carsten

Excellent. I run Linux so Scilab would be the choice. And the 5 lines of code would be???



Cheers

Ian

 

[culteousness1]

Unfortunately, I can't test it right now as I am on my dayjob.
But IIRC you will need at least some I/P handling for the wav file:

[x,sf,bits]= wavread('testfile.wav');

Next you can apply the FFT:

X = fft(x);

You will need a frequency vector f to plot the result:

plot(f, abs(X));

I probably forgot some essential things, so better have a Google
to fill the two remaining lines of code 

HTH,
Carsten

 

[ruffrecords]

Hi Ian,

Are you on Mac or PC?

Can you email me the sweeps or a link to them?  I can try to import to Fuzzmeasure.

Cheers,
Ruairi

I am on a PC. The files are in the form of a 270M+ zip file.  I think I can email you a link to them.

Cheers

Isn

 

[ruffrecords]

Unfortunately, I can't test it right now as I am on my dayjob.
But IIRC you will need at least some I/P handling for the wav file:

[x,sf,bits]= wavread('testfile.wav');

Next you can apply the FFT:

X = fft(x);

You will need a frequency vector f to plot the result:

plot(f, abs(X));

I probably forgot some essential things, so better have a Google
to fill the two remaining lines of code 

HTH,
Carsten

I am struggling. The Scilab documentation is sparse so getting the arguments right has been a pain. What I have so far is this:

stacksize('max');

[y, Fs, bits]=wavread('/home/ian/Music/610sweeps/Sweeps/High/4.5khz/Boost/610_4k5_+3db.wav');
printf("%d\n",Fs);
printf("%i\n",bits);

analyze(y, 20,20000,Fs);

The stacksize was necessary to cope with the size of the data. I could not get fft followed by plot to work but analyze is supposed to plot an audio file but the plot looks nothing like what I expected (see attached).

Cheers

Ian