Official 33609 builder's tread. See 1st page for updates.
- Thread starter Igor
- Start date
Help Support GroupDIY Audio Forum:
bradb
Well-known member
Igor,
would you consider offering this 33609 in the future as a KIT? all parts included?
would you consider offering this 33609 in the future as a KIT? all parts included?
ultra-alex
Well-known member
rob_gould said:Hi Alex,
There's loads of info on the different sounds of different resistor materials. Just google and you'll turn up loads of hits.
On the subject of values / power / tolerances, it depends on the situation. For the switches on the 33609, you'll find you need 1% tolerance because of how specific some of the values are. 1/4 watt are fine for this application.
Generally speaking, 1% 1/4 watt metal oxide are what I use, and this seems to be common for a lot of people too, unless the BOM specifies a higher power rating.
Often a BOM will call for 5% tolerance for resistors, but they're so cheap it's probably better to build up a collection of 1% anyway...
It may also be useful for you to read this articale about the standard resistor values:
http://www.logwell.com/tech/components/resistor_values.html
Hope this helps,
Rob
it helped! btw, beautiful work, rob!!!!
i just recieived my pcbs two days ago and i´am totally stuck in my (little) lab ever since...
i already finished the two main and the power pcbs. so the next question will be the choice of transformers.
i´am pretty shure that there will be a lot of questions to come, so it´s a qood feeling that i´am not alone here!!!
cheers, alex
p.s. igor, i don´t want to bother you with my stupid questions. if so, don´t answer.
and this is not meant disrespectful! my deepest respect for your dedication, your work and of course your knowledge!
p.s.s. thank you, tat! honest and straightforward...
ultra-alex
Well-known member
some questions:
the c5 on the BA340 below is 10nF ceramic, right?
how do you guys connected BA340 with main pcb? just solder 1,2,3,5,6?
and can anybody clarify on the 960OP option. is that instead of the BA340? if not, what is it for
psu: if i´am using R5 i don´t need R3 trimpot, right?
what exactly does the "for adjustment only" area in PSU pcb? is it just an additional DC source?
i´ve read through the whole thread at least two times. sorry if any of my questions have already been discussed and i´ve overlooked it.
and thanks in advance for any help...
the c5 on the BA340 below is 10nF ceramic, right?
how do you guys connected BA340 with main pcb? just solder 1,2,3,5,6?
and can anybody clarify on the 960OP option. is that instead of the BA340? if not, what is it for
psu: if i´am using R5 i don´t need R3 trimpot, right?
what exactly does the "for adjustment only" area in PSU pcb? is it just an additional DC source?
i´ve read through the whole thread at least two times. sorry if any of my questions have already been discussed and i´ve overlooked it.
and thanks in advance for any help...
rob_gould
Well-known member
ultra-alex said:
I'll have to check this when I get home. Doesn't the BOM give you a clear idea?
use mill max pins or something similar:
http://www.mill-max.com/
I don't have a part number, but I did manage to order 50 as a sample request from their website.
I don't recall I'm afraid. Can check later though.
yep that's right
IIRC it's for use during the calibration procedure. You need to build it, but it's not for use after the unit has been completed and calibrated.
ultra-alex
Well-known member
i just richards BOM. and there it says that 10nf for BA340 is C4. i can´t find a C4 on 340board, only the C5 on the bottom...
thanks for your help rob!!
thanks for your help rob!!
I used regular snap-off IC socket pins like these ones. The pins will quite happily stack on top of each other piggyback-style.
Snap one pin off and clean the black plastic off it. Install it on the main board so that the socket part points upwards. Snap another pin off and clean the black plastic off. Install it on the BA340 board so that the pin part points downward. Apply solder to the top side so that excess solder doesn't get in the way of the pin (from memory it was easier to install the pins in the BA340 before installing the other components). Repeat for all sockets/pins. The pins/sockets will hold the BA340 tight enough to prevent it falling out, and you can still remove it later on if you need. Also a lot cheaper than the Mill-max pins.
ultra-alex
Well-known member
first of all, i should really start reading a lot more concentrated!!
second, i should visit a "how to crimp properly" workshop! i hate molex...
second, i should visit a "how to crimp properly" workshop! i hate molex...
Hi Alex, 990 option is to use 990 style discrete op-amp instead of ba340.
There's components marked Cj, Rj, etc, at bottom silkscreen. Don't install them with 340.
They go only with 990 style opamp, to bias input to halfsupply and support feedback path.
IIRC nobody wanted to use 990 option, and it died on my proto, but tested.
Molex: put crimp pins by half (not till the end) into housing and solder wires.
Wire should be pre-covered with solder, it simplifies process.
It is possible to find some video howto solder crimp pins on Youtube imho.
There's components marked Cj, Rj, etc, at bottom silkscreen. Don't install them with 340.
They go only with 990 style opamp, to bias input to halfsupply and support feedback path.
IIRC nobody wanted to use 990 option, and it died on my proto, but tested.
Molex: put crimp pins by half (not till the end) into housing and solder wires.
Wire should be pre-covered with solder, it simplifies process.
It is possible to find some video howto solder crimp pins on Youtube imho.
RuudNL
Well-known member
Hi!
I just found this forum after I created my own 33609 replica...
In fact it is not exactly a 33609, but a slightly modified and improved version.
Here are the most important changes:
- I left out the intermediate stage transformer and used a differential amplifier instead.
- The range of compress and limit threshold is somewhat modified to suit my needs,
- I have added a 5:1 ratio to the compressor section.
- Gain makeup is in 1 dB steps instead of 2 dB steps.
- Input and output transformers were robbed from old broadcast equipment, they were custom made and are high quality.
- Front panel (4 mm.) is engraved and coloured in.
- Meters are Sifam presentor. (Farnell)
- Knobs are Mentor 20 mm. (Conrad)
I have used a small bridge rectifier as a control element. (B40C1000)
The advantage is that all diodes are in the same enclosure and always have the same temperature.
Matching of the individual diodes is better than 0,5 mV. I have selected the best pair out of 20.
(By the way: Is there somebody who has the original selection criteria for the diodes in the attenuator? Because not only the matching is important, but also the characteristic I assume.)
Specs are better than the original Neve specs, especially with respect to the noise floor and distortion figures.
And it sounds great! I used to work as a recording engineer and always liked the Neve compressors.
Sometimes you saw the needle pointing around 6 dB gain reduction (or higher...), but you didn't hear the action of the compressor!
This reproduced unit gives me he same feeling.
Just as a joke ("designed by Rupert Neve") I added my own name on the front panel...
Great to read that there are so many enthousiasts who are trying to keep the 33609 alive in their own way!
http://www.ljfaahw.nl/BBC/compressor.jpg (2066 x 942 px. 175 kB)
I just found this forum after I created my own 33609 replica...
In fact it is not exactly a 33609, but a slightly modified and improved version.
Here are the most important changes:
- I left out the intermediate stage transformer and used a differential amplifier instead.
- The range of compress and limit threshold is somewhat modified to suit my needs,
- I have added a 5:1 ratio to the compressor section.
- Gain makeup is in 1 dB steps instead of 2 dB steps.
- Input and output transformers were robbed from old broadcast equipment, they were custom made and are high quality.
- Front panel (4 mm.) is engraved and coloured in.
- Meters are Sifam presentor. (Farnell)
- Knobs are Mentor 20 mm. (Conrad)
I have used a small bridge rectifier as a control element. (B40C1000)
The advantage is that all diodes are in the same enclosure and always have the same temperature.
Matching of the individual diodes is better than 0,5 mV. I have selected the best pair out of 20.
(By the way: Is there somebody who has the original selection criteria for the diodes in the attenuator? Because not only the matching is important, but also the characteristic I assume.)
Specs are better than the original Neve specs, especially with respect to the noise floor and distortion figures.
And it sounds great! I used to work as a recording engineer and always liked the Neve compressors.
Sometimes you saw the needle pointing around 6 dB gain reduction (or higher...), but you didn't hear the action of the compressor!
This reproduced unit gives me he same feeling.
Just as a joke ("designed by Rupert Neve") I added my own name on the front panel...
Great to read that there are so many enthousiasts who are trying to keep the 33609 alive in their own way!
http://www.ljfaahw.nl/BBC/compressor.jpg (2066 x 942 px. 175 kB)
rob_gould
Well-known member
WOW!
Beautiful work.
Beautiful work.
crisotop
Well-known member
very nice indeed!
Hi RuudNL , looks nice from outside, but only thing can impress me is good inside built.
You can match diodes by Vf using test fixture like dip-8 socket,
switch x4 pos (to match 4 diodes at same time), 5v source, spdt switch, 2k and 20k resistors.
Write down Vf values with 2k and 20k, that's all.
After putting diodes in dip8 socket, wait 20-30 sec to let them get same temperature.
Regarding interstage trunny, it is very important part of sound. I'd not tried to save here few bucks.
Interesting thing, but when needle at 6db, actually gain reduction is 8-10 db at peaks.
Sifam are very slow meters, they can't react so fast.
Hm...you really can't hear it?
Another little thing. Interstage trafo not only amplifies the signal after diode bridge, but matches impedances.
It is not big deal to get bit better specs as specified in test procedure, it has some tolerance;
but have you compared it side-by-side with original unit?
(just curious, I tried few differential amps instead of interstage, including INA103 and ssm's, but found
it less "sounding" to my bad taste).
You can match diodes by Vf using test fixture like dip-8 socket,
switch x4 pos (to match 4 diodes at same time), 5v source, spdt switch, 2k and 20k resistors.
Write down Vf values with 2k and 20k, that's all.
After putting diodes in dip8 socket, wait 20-30 sec to let them get same temperature.
Regarding interstage trunny, it is very important part of sound. I'd not tried to save here few bucks.
Interesting thing, but when needle at 6db, actually gain reduction is 8-10 db at peaks.
Sifam are very slow meters, they can't react so fast.
Hm...you really can't hear it?
Another little thing. Interstage trafo not only amplifies the signal after diode bridge, but matches impedances.
It is not big deal to get bit better specs as specified in test procedure, it has some tolerance;
but have you compared it side-by-side with original unit?
(just curious, I tried few differential amps instead of interstage, including INA103 and ssm's, but found
it less "sounding" to my bad taste).
RuudNL
Well-known member
Hm...you really can't hear it?
Well, of course I meant no audible 'pumping' or other audible side effects...
I have built the compressor on two home made PCB's, one containing the audio part and the second containing the compress and limit side chains.
I will see if I can make a picture of the inside.
The response time of the Sifam meters is acceptable for me, I do not consider them as 'slow'.
Anyway: knowing the sound of the original Neve compressors, I am very happy with the result.
Sure. This is great thing and good to have one in studio.
Pumping can be used as sound effect and not always unwanted thing, this is why total cleaning
the artefacts will make different character compressor.
Sifams are acceptable to me as well, just wanted to note about their actual responce.
Don't pay too much attension to my way of kidding, btw
Pumping can be used as sound effect and not always unwanted thing, this is why total cleaning
the artefacts will make different character compressor.
Sifams are acceptable to me as well, just wanted to note about their actual responce.
Don't pay too much attension to my way of kidding, btw
ultra-alex
Well-known member
Curtis said:
thanks, i´ll try this! the last two hours i tried to build my own "340 standoffs". again not quite successful...
ultra-alex
Well-known member
a short question regarding the SPDT switches with centre off. are these right or not: CK7103, SPDT, ctr off, farnell 9575049.
and i can´t find the elma 04-1130-20 with printcontact anymore!? can i go with the elmas 04-1133/04-1130?
thanks in advance, alex
and i can´t find the elma 04-1130-20 with printcontact anymore!? can i go with the elmas 04-1133/04-1130?
thanks in advance, alex
rob_gould
Well-known member
Nah they're not the right ones. The centre off switches were a mistake on the bom. The correct switches are 2 position. Try 9575030
With regard to the Elmas, I'm not sure exactly what you mean by your question, but try Ptwonkid's online store. diypartssupply.com I think. I'm pretty sure he has the right ones.
Also, did you sort out the ba340 pins? I can stick some in the post to you if you still need them. Sorry if it's too late. I haven't checked this thread for a few days.
With regard to the Elmas, I'm not sure exactly what you mean by your question, but try Ptwonkid's online store. diypartssupply.com I think. I'm pretty sure he has the right ones.
Also, did you sort out the ba340 pins? I can stick some in the post to you if you still need them. Sorry if it's too late. I haven't checked this thread for a few days.
nielsk
Well-known member
Is anyone else having meter problems? GR is working properly, but the meter will not turn down from slammed.....
there are a few value changes in different revisions,
R32 being 390 or 680
R35 being 7k5 or 10k
the meter is a Sifam M109760/2
if I lift R35 I get adjustment..
there are a few value changes in different revisions,
R32 being 390 or 680
R35 being 7k5 or 10k
the meter is a Sifam M109760/2
if I lift R35 I get adjustment..
nielsk
Well-known member
I must be missing something obvious, it looks like the 10k trim pot should be able to bring the meter down to nothing....I have allready replaced the pot, BTW
all components look to be correct..
all components look to be correct..
Similar threads
