sontec 1 RU build thread

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Hi ,

First of all Thanks Igor, great project!
Quick question. I've seen lots of great talk on the 4 pin molex, but I have a question on the 5 pin molex, I see from pcb and pic the spacing is smaller. I just want to be sure before I order is it the .1" spacing for the 5 pin?
Thanks!
AC
 
Hopefully, logged as Igor :)
4-pin connector has 3.81mm spacing
5-pin connector has 2.54mm spacing
In case crimping/soldering the wires to Molex pins seems hard, another opt:
http://il.farnell.com/jsp/level5/module.jsp?moduleId=en/230246.xml
this is for 2.54 mm,
for 3.81:
http://il.farnell.com/phoenix-contact/1727036/terminal-block-pcb-3-81mm-4way/dp/3704592

I personally hate screw terminals, but seems, it is same thing as Word Cup :)

POST Nr 1272!!!!!!!
 
My boards for my second sontec came in yesterday and this one is going to be a bit more special.
It will be a 3u version with all stepped controls (elma's i think.. though very expensive), forsell 993's and it will also have left controls on the left and right controls on the right.
I was thinking of cutting the filterboards in 2 or maybe even 5 parts so that i can mount them vertically next to the switches.
This way every secion will have it's own filterboard and practically the board can 'handle this', it even has a mark where to cut it on the bottom.  ;)
I will only have to connect a couple of wires to make it work, basically psu, ground and busses so fast thinking 6 wires per board.

Just wondering, will i run into other poblems then just some extra wiring? Don't think so myself but want to make sure.
Personally i think it might be a good idea to have the summing together near/on the IO boards.
 
OK, have been having a better look at the board and it's not real easy to cut the board in 5 pieces but at least doable in 2 pieces.
You can cut in 2 between the LMFfreq and MQ pot, exactly on the IJ Research logo (sorry Igor).
You can also cut the HF section apart, there is only a 10k resistor that has to be taken care of.
OK, we'll see..First find a nice source for a lot of elma's or equivalent. (anyone?)
 
zayance said:
I bought 20 chinese stepped attenuator on ebay for my Sontec the dealers name is hampton if i remember correctly, got me a good price for them, they are Vishay i believe.
Seems like we will be building same sontec, i went for forsell also ;-)
Please tell me more! (concerning the switches that is) :)
 
@ Rednoise: Good for you, you have more posts :)
Cutting my logo!!!! Will never forgive you!!!  :p  ;D
Regarding the ELMAS. I did some boards to solder resistors on PCB,
not to ELMA contacts. Can post a picture here. Useful.
If there are too many people who wants mastering version...
Let's see what is possible :)
 
dagoose said:
zayance said:
I bought 20 chinese stepped attenuator on ebay for my Sontec the dealers name is hampton if i remember correctly, got me a good price for them, they are Vishay i believe.
Seems like we will be building same sontec, i went for forsell also ;-)
Please tell me more! (concerning the switches that is) :)

let me get back to my desk and i'll send you the ebay link :)
 
Thinking about the stepped switches; for frequency i'm going to use the calculations as vitalahde describes so with semitones instead of regular log. Has anybody done that already?
For the Q i will go for 12 steps switches because i think that is more then enough and for the gain i will also use 24 steps switches.
Both Q and gain are lin so i guess it's just a matter of deviding it in all the same values, so for gain it's something like 2k per step x 24 steps = 48k (almost 50k) and the centre will also be 0 automatically. Am i correct here or do i miss something?
Elma's will be a bit to much i guess so i will probably go for the switches from uraltone, about 1/10 of the price.
 
dagoose said:
Thinking about the stepped switches; for frequency i'm going to use the calculations as vitalahde describes so with semitones instead of regular log. Has anybody done that already?
For the Q i will go for 12 steps switches because i think that is more then enough and for the gain i will also use 24 steps switches.
Both Q and gain are lin so i guess it's just a matter of deviding it in all the same values, so for gain it's something like 2k per step x 24 steps = 48k (almost 50k) and the centre will also be 0 automatically. Am i correct here or do i miss something?

Anyone?  ???
 
dagoose said:
dagoose said:
Thinking about the stepped switches; for frequency i'm going to use the calculations as vitalahde describes so with semitones instead of regular log. Has anybody done that already?
For the Q i will go for 12 steps switches because i think that is more then enough and for the gain i will also use 24 steps switches.
Both Q and gain are lin so i guess it's just a matter of deviding it in all the same values, so for gain it's something like 2k per step x 24 steps = 48k (almost 50k) and the centre will also be 0 automatically. Am i correct here or do i miss something?

Anyone?  ???
I only did the calculation in the excel sheet, dunno about Jaakos calculation. It's simply his idea translated into numbers, his values probably not that different from mine.
For your 24 steps and center, sure about 24 steps (does something like this exist?), not 24 positions? else you'll be missing the center position for +/-0dB. All/most 24 position switches, like Uraltone or Chiefdom RCL 370, other than Elma series04, don't come with end stop, that could kill yor monitors when dialing from electrical full ccw position one step further ccw to full cw. The ones I know that come with end stop or don't dial thru only have 23 positions/22 steps for 15° angle of throw. 
 
Re stepped gain controls: I got around this by using 24 step switches with steps 1&2 tied together. That way you can't have the switch 'fall off the end' into an unconnected step. The center tap point is then step 13 and you have 11 steps either side of this (actually 12 on the 'anti-clockwise' or minus side but 2 are tied together with a zero ohm link). Done this way, you need 22 x 2k2 resistors giving 48k4 total which is only 3.2% off 50k.

What's really needed is a 25 position switch as then you'd have 12 positions either side of zero (the center tap point at step 13), giving +/- 12dB nicely. Or of course, use actual 23 position switches giving +/-11dB.






 
thanks! i was thinking in the right direction with the gain switches. :)

I just ordered 20 switches from uraltone this afternoon, they are indeed 24 steps but it's really easy to make 'm 23 steps which i'm going to do.
All you have to do is drill a 1.5mm hole in the plastic and put in a metal pin and it's a 23 steps switch, done this a couple of times before, needs some practice but now i know where to drill.  ;)
I will make a pic of the place wher to drill a hole and put the pin in, hard to explain, you have to see..

23 steps will also give the 0 position at the centre indeed, nice! Now it's time to order a lot of 2k2 resistors and start soldering..  :-*
I'm still thinking about the freq switches but that will also work out well i think. Can't wait to hear and work with my stepped DOA sontec.  ;D


 
MrShhh said:
Re stepped gain controls: I got around this by using 24 step switches with steps 1&2 tied together. That way you can't have the switch 'fall off the end' into an unconnected step. The center tap point is then step 13 and you have 11 steps either side of this (actually 12 on the 'anti-clockwise' or minus side but 2 are tied together with a zero ohm link). Done this way, you need 22 x 2k2 resistors giving 48k4 total which is only 3.2% off 50k.

What's really needed is a 25 position switch as then you'd have 12 positions either side of zero (the center tap point at step 13), giving +/- 12dB nicely. Or of course, use actual 23 position switches giving +/-11dB.

Then you can truthfully say that youe EQ goes to "11"  ;-)
Bruno2000
 
Back
Top