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

[dmp]

Can anyone do me a favor and confirm wether this DC offset is supposed to be there
and their on their build. You would have to measure at the interstage primary.

I measured about 4.2 volts on the primary of the interstage transformer tonight as I put mine back together.

As I was trying it out tonight, R44 blew and the output stopped. I had used 1/4w - but with ~300 mA through a 10 ohm resistor, the steady power is .9w, so R44 and R43 should both be at least 1watt I think. Swapped these out and it is working again. For some reason one of the BC461 blew at the same time (these are available from Newark btw)

I have an example of compressed drums now through my unit... any suggestions on a good way to post it? it seems like most people use there own servers?

 

[dmp]

Here's an example on a drum bus. Pretty heavy compression and I don't hear unpleasant distortion. I have R10' = 6K8. About 7dB of reduction on the snare peaks.
Settings: Limiter +4,50ms,fast / Compressor -12,100ms,G=6,4:1
There's a little bleed from other instruments from the original recording session. The only other effect is a lexicon reverb plugin on the snare.
http://www.probstinnovation.com/Drums2_33609b.mp3
http://www.probstinnovation.com/Drums2_bypass.mp3

 

[Biasrocks]

So you do have voltage across the primary of the interstage, I wonder if it's because there's no ground reference and we're just not measuring correctly, because I can't see anyway that voltage would creep in there.

I've got a 1/4W in that position, adjusted for 180mv (iirc) across and it doesn't seem to get hot under use. If you have the bias set to high on the BA340, you'll definitely go through a few of those 10R's. I think when you're building, it's good to use a 1/4w'er there to avoid biasing the output stage to hot. Once you get it settled in a 1/2w'er is probably not a bad idea.

Listening to your examples, you may want to consider using 24K in the R10' position. I think it's better for all around use, and you still have the limiter section @ 6K8 to catch the peaks if you need to. I found using the limiter in the fast position caused some problems down the line when trying to processing it again. For instance I used the fast attack on a mix, and it consistently caused problems with the Sonnox Limiter distorting quite badly with very conservative settings. I switched to slow attack and the problem went away, so be careful with that. I think Neve didn't recommended using the fast setting for music applications.

As for the distortion pointed out in the previous posts, I haven't experienced any issues. If you look at the settings that were used what did you expect??? No one in their right mind is gonna use a 6:1 setting on a mix pulling 20db+ of gain reduction, that's just asking for trouble.

Those Elma's feel nice don't they?

Mark

 

[mg73]

Huh?
The BC441/BC461 is a complementary pair used to drive the output, like push-pull. They need to be npn / pnp to work correctly, so changing

them to 2n4037

is not going to work.

2N4037 is a common replacement for Bc461 and work fine!

 

[dmp]

Power=I^2*R = .18^2*10=.324Watts. I think you are still overloading a 1/4w resistor. If the sound cuts out you'll know where to look.
Igor recommended biasing to 260-280mV I think. Mine was 240mv with the trimmer fully ccw, so I don't know how you got 180mv unless you used different components.
I agree about the distortion earlier in the thread - I ran that test and the meter was slamming into 20dB, which causes distortion in most compressors, I think.
I measured the release with the 6.8k resistor and it was right around 100ms. Would you be interested in running the same file through your unit with the 24k resistor? I'd be interested to hear how they compare.
The elma's are really nice. I think having them on the compressor threshold is really worth it - I haven't gotten them for the limiter side, but it would be nice since the compressor gain is in such coarse steps.

 

[dmp]

I misunderstood you since

them

is a plural pronoun (..referring to BOTH BC441 and BC461). I was trying to helpfully point out that the same part cannot be a common replacement for BOTH.

 

[mg73]

I misunderstood you since

them

is a plural pronoun (..referring to BOTH BC441 and BC461). I was trying to helpfully point out that the same part cannot be a common replacement for BOTH.

it´s two BC461/2N4037 on a BA340

 

[Biasrocks]

Igor recommended biasing to 260-280mV I think. Mine was 240mv with the trimmer fully ccw, so I don't know how you got 180mv unless you used different components.

Right, 280mv is where I ended up, I mis-remembered

I can't feel any heat build up on the 1/4w R10 but it's probably still a good idea to go up to 1/2w there.

I measured the release with the 6.8k resistor and it was right around 100ms. Would you be interested in running the same file through your unit with the 24k resistor? I'd be interested to hear how they compare.

Sure, send me a .wav file and I'll run it through.

Mark

 

[dmp]

Here is the raw input file:
www.probstinnovation.com/Drums2_raw_24bit.wav

I also ran with the Limiter in slow... but can't hear much of a difference
www.probstinnovation.com/Drums_33609b_LimSlow.mp3

 

[dmp]

So you do have voltage across the primary of the interstage, I wonder if it's because there's no ground reference and we're just not measuring correctly, because I can't see anyway that voltage would creep in there.

I think it must be leakage of C1/C2, and because there is no drain on the primary of the transformer, it shows the voltage of the BC214 emitters upstream. Still, I am surprised to see this also.

 

[Biasrocks]

I also ran with the Limiter in slow... but can't hear much of a difference

Most likely because you have the compressor clamping down at the same time as the limiter because
of the 6K8 in the compressor section.


Here's the processed files.

Drums 6:1, Rel:100ms, Limit: Attack: Slow, Release 50ms

http://sharktankpro.com/GDIY/33609/Drums-6to1.wav

Drums 4:1, Rel:400ms, Limit: Attack: Slow, Release: 100ms

http://sharktankpro.com/GDIY/33609/Drums-4to1.wav

First thing I noticed is the attacks seem much more prominent.

Mark

 

[dmp]

I'm confused because the 6.8k is on the release... are you saying a higher resistance (which increases the release time) will create a more prominent sounding attack?
Thanks for running the files - I'll have a listen tomorrow.
(the second one has a typo I think, should be Drums-4to1.wav)

 

[Biasrocks]

Must stop typing late at night....

Mark

 

[Wonderlandaudio]

Correct resistors are 6k8 in the limiter and 24k in the compressor...

 

[dmp]

Mark,
I think hear what you are talking about - especially audible on the kick drum - kind a harshness at the start. Trying slow attack today to get rid of that and I like the results more. Thanks much.
I'm still not sure on switching the 6.8k to 24k. I do hear a slight difference but I kind of like the edge when I ran the same 6:1 settings through my unit as you did. It may be that I am getting the "unpleasant" distortion and I'll figure that out eventually. I hear it on the kick drum especially. I have the 1173 output transformer however, so our builds have other differences besides R10'.

www.probstinnovation.com/Drums-6_1_dmp.wav
Not sure if I matched the level of reduction you were getting, but this was about 6-8 dB on the peaks

 

[Biasrocks]

I wasn't pulling quite as much gain as you, about 4-5db max so I redid it pulling more.

http://sharktankpro.com/GDIY/33609/Drums-6to1-ReDone.wav

Mark

 

[Mirouf]

Hi to all!

I'm not making a 33609 (for this time...) but I'm using some of Igor 24V power supply, and as I don't find schematics nor explaination about them, I'm wondering a few things:

- I notice (on a picture) that the trim pot isn't needed if you fit the R5(3K57) resistor, Am I right? But If I use the trim pot I don't need R5, but in both case I need R1(196R)?

- What is the "For adjustement purpose only" Section for?Do I need to populate this place, or is it a feature for 33609/2254?

Thanks!

 

[Rob Flinn]

You are right that the trim pot isn't needed if you use the 3k57 resistor.  But the 196R is needed.

The "adjustment purposes section" is a fixed low impedance voltage reference which is needed when you calibrate the Neve compressor/limiters.  You can use pretty much any single DIL op amp in there, because it is only connected to the compressor when you calibrate the unit so won't affect the audio quality.

I think all this is actually covered in this trhead before if you go back & look through it.

 

[Mirouf]

Thanks! I've got my answer

Sorry, but the 40 pages it's difficult, even with "search" function,  to find something without spend 5 hours 

 

[Biasrocks]

Sorry, but the 40 pages it's difficult, even with "search" function,  to find something without spend 5 hours 

I always read the entire build thread before starting a build, and will reference it often throughout.

Time well spent.

Mark