Official 33609 builder's tread. See 1st page for updates.

[kazper]

Hi guys,

in Igor's 2254/33609 Tekay mentions this issue

http://www.groupdiy.com/index.php?topic=312&postdays=0&postorder=asc&highlight=2254&start=75


To quote him -

"Just found a wrong value fault in the 33609 manual regarding the Compressor Recovery switch.
The 6k8 resistor shall be 24k not 6k8 as shown in the EX20236 drawing!
With 6k8 the fastest recovery will be around 25mS!"

Thanks Tekay!!

Drawing PL20235 now?

 

[keithcamilleri]

hey igor justr a quick question...

both boards are working fine ...one meter when i switch on stays up for 2sec and the other one goes up and straight down....is there something i need to check...?

thanks

keith

 

[rob_gould]

Hi Igor,

What's the approximate current output of the power supply for this project? 

I'm trying to calculate the resistor for the lamps on meters...

Thanks!

Rob

 

[Igor]

both boards are working fine ...one meter when i switch on stays up for 2sec and the other one goes up and straight down....is there something i need to check...?

Hm... If boards passed test procedure 100%, should be OK.

What's the approximate current output of the power supply for this project?

1A, 24V. I prefer LED's for backlighting. Less current, never blows out.

 

[rob_gould]

[quote author=Igor]

What's the approximate current output of the power supply for this project?

1A, 24V.
[/quote]

So i want each bulb to drop about 10v rather than their full 12v rating for extended life.  This means the resistor has to drop 4v right?

I have calculated that if the power supply gives 24v 1a, this means I need a 4ohm 4watt resistor in series with the two bulbs. 

Does this sound correct?

Cheers,

Rob

 

[Rob Flinn]

[quote author=Igor]

What's the approximate current output of the power supply for this project?

1A, 24V.

So i want each bulb to drop about 10v rather than their full 12v rating for extended life.  This means the resistor has to drop 4v right?

I have calculated that if the power supply gives 24v 1a, this means I need a 4ohm 4watt resistor in series with the two bulbs. 

Does this sound correct?

Cheers,

Rob
[/quote]

The psu supplies what is requested of it.  It doesn't "give" whatever it is capable of supplying to any circuit that happens to be hooked up to it.  The amount of current drawn is a function of the resistance of the circuit connected across it.

If you bulbs were drawing 1A at 12v each then they would be 12watt bulbs (P=IV) which I suspect is unlikely for backlighting a meter.

If you are connecting 2 bulbs in series across the 24v the combined resistance of the series bulbs will determine the current being drawn.  You can't just stick a meter across the bulbs to work out the resistance because the bulbs resistance changes when they heat up.  Therefore you might want to take the bulbs power rating & using the equation P=IV you can calculate a rough current draw that will enable you to work out the dropper resistance value you require.

But like Igor says I'd be using led's if I were you cos they don't blow so easily.  ;-)

 

[rob_gould]

[quote author=Rob Flinn]
But like Igor says I'd be using led's if I were you cos they don't blow so easily.  ;-)
[/quote]

Hehe thanks for the info Rob.

It seems like the general consenus is definitely that using the bulbs is not as good as using LEDs, but having paid for them, this time I'm going to use them

I'll know for next time though...

 

[Curtis]

Hi chaps,

Everyone's probably sick of seeing these now, but I've finally finished my unit.  All calibrated and sounding absolutely magic  8)

Another angle:

http://i153.photobucket.com/albums/s222/ac1176/3360903.jpg

Topless (phwooooaaarrr!):

http://i153.photobucket.com/albums/s222/ac1176/3360902.jpg

The trannies are all Carnhill BTW.

No real dramas to speak of.  Only real problem I had was I was supplied a mixture of BC214L's and BC214C's (different pinout), and had to make sure I installed the L's with a "twist" in the legs - got one transistor mixed up and installed a C with the L's pinout.

I think someone earlier mentioned they had problems with hum?  I found that I got the quietest operation by linking the two grounds on the PSU together.  There's a exposed rectangular copper pad on the PSU just near the heatsink which is the ground for the XLR pin 1's.  I found that if I linked this pad to the nearest circuit ground it got rid of the last of the residual hum.

Picture:

http://i153.photobucket.com/albums/s222/ac1176/33609gnd.jpg

Thanks to Igor and Purusha for making such an excellent project available  ;D

 

[RedNoise]

Gorgeous!!!
Very clean inside!!!
thanks for the pics! ;D

 

[maxheadroom]

looks very good inside!!

have you made a RMAA measurement?

 

[Igor]

@Curtis:
Cool, man!  I very happy how it looks inside!

 

[Curtis]

Thanks chaps

have you made a RMAA measurement?

No, sorry.  I could possibly do a couple of noisefloor measurements with the DAW and soundcard if you want?

The hiss level in mine seems quite alright.  Like Igor said, it's not the quietest compressor out there, but provided you're not running it with the gain makeup control maxxed out it's fine.

 

[maxheadroom]

Thanks chaps

No, sorry.  I could possibly do a couple of noisefloor measurements with the DAW and soundcard if you want?

that would be fine!!!

 

[Curtis]

I did some quick and dirty measurements last night.  I've got some screenshots of the spectrum plots from the plugin I used, but I'm stuck behind a firewall here at work and can't get them to link properly.  I can post them later on this evening from home if you like.

Long and the short of it was I was seeing about -58dB noise in both the left and right channels with all the controls set fully CCW, except for the gain makeup control which was set all the way up (+20dB).  Stereo Link was on, Bypass In, Comp and Limit for each channel In, Limit set to Fast.  Nothing connected to the inputs.  Soundcard interface set to 0dB (Presonus FP10).  I doubt that the results I got were completely scientifically accurate, but they should give you some indication of what I'm seeing in my unit.  With the unit set to a more respectable makeup gain level the noise disappears into the background.

Page two of the test procedure suggests that the nominal noise figure you can get is better than -75dB with the makeup control set to minimum, and -55dB with the makeup gain all the way up, so I guess the figures I'm getting are pretty much the same as the original (assuming my test method is OK).

Edit: just re-read some of your earlier posts - I didn't know RMAA was free, I'll download it tonight and retest my 33609 using it so you can compare the results properly.

Just had a quick look at the Neve website.  They quote their 33609JD (their modern repro) as having -55dB noise figure with the gain makeup control at +20dB, so maybe this is as good as it gets?

 

[Curtis]

Spectrum plots from my DAW. Left channel:

Right channel:

http://i153.photobucket.com/albums/s222/ac1176/33609noiseR.jpg

I'll re-test using RMAA and post the results.

 

[Curtis]

OK, RMAA says:

Left channel RMS noise = -73.7dB
Left channel RMS noise (A weighted) = -77.7dB
Left channel Peak level, dB FS = -60.3dB

Right channel RMS noise = -72.5dB
Right channel RMS noise (A weighted) = -76.2dB
Right channel Peak level, dB FS = -57.8dB

That's with limit/comp In, bypass In, gain at max, all other controls fully CCW, stereo link Off.  So I guess my previous test with the DAW was OK (-58.3dB/-59.5dB peak noise vs -57.8dB/-60.3dB peak noise).

 

[Igor]

Looks 100% up to specs.

 

[maxheadroom]

thank you very much curtis!!!!
are these values in dbFS ?

 

[Curtis]

Yep, dBFS.

 

[maxheadroom]

when your setup is like 0dbFS = +18dBu then you must decrease 18db from your values to compare them to the neve specs

EDIT: i ment +18dBu