New 'ultimate' SSL buss comp clone ;-)

[khstudio]

I spoke with Keith for a while tonite & it seems the instructions do have a mis-print :!:

9. Connect “C” & “D” on the slave main pcb to “C” & “D” on the slave switch pcb.

Keith said that C & D should connect from Master to Slave PCB... Not to the Slave switch PCB.

Rob... I DID use the files & NOT the pics to figure this out & noticed that pin "C & D" on the switching PCB goes nowhere, so it made no sense.

I appreciate your help man & don't mean to be a dick but please check the instructions for errors... I'm confused.

Also,
Keith said that pins A & B on the slave switch PCB should connect to the Slave Main A & B... & I assume (since it's on the instructions) that A ALSO needs to connect/jump over to the MAIN PCB A :?: I read the entire thread and never saw metion of the "A"s connecting on the two MAIN PCB (Just the switch PCB) so what is up here???

The A & B don't make sense... shouldn't the channels stay separated until AFTER the detector = @ point "C" - where they join together???

Kevin

 

[khstudio]

Include the TL072 in the slave build, do not omit it as I previously tried... -This way works better.

Wow... I was looking at the TLO72 section on the slave & thought it was a mistake on the last set of instructions until I found Keiths previous notes on page 2. This is very strange that this "section" is needed twice :shock:

You wouldn't think so lookin at the schematic & the way the path is going - with only one Mak-up & threahold control... but OK, I'll do it.


BTW,
Thank you Keith & Rob for helping me thru this
:guinness: :guinness: :guinness: :sam: :sam: :thumb:
... I'm just trying to get a FULL grip on this to make it happen, correctly.

Kevin

 

[Rob Flinn]

[quote author="khstudio"]I spoke with Keith for a while tonite & it seems the instructions do have a mis-print :!:

9. Connect “C” & “D” on the slave main pcb to “C” & “D” on the slave switch pcb.

Keith said that C & D should connect from Master to Slave PCB... Not to the Slave switch PCB.[/quote]

Points "C" & "D" are still points "C" & "D" whether you make the link on the switch pcb or the main pcb, since they are actually connection points between the master & switch pcb's. i.e they exist on both master & switch pcb.

Keith said that pins A & B on the slave switch PCB should connect to the Slave Main A & B... & I assume (since it's on the instructions) that A ALSO needs to connect/jump over to the MAIN PCB A Question I read the entire thread and never saw metion of the "A"s connecting on the two MAIN PCB (Just the switch PCB) so what is up here???

The A & B don't make sense... shouldn't the channels stay separated until AFTER the detector = @ point "C" - where they join together???

Pins A & B should be hooked up from main pcb to switch pcb's individually for each set of pcb's. This is because the ratio is part of the rectifier circuit & if you didn't make this connection then the slave would have no ratio control. Also the resisitors & diode that you have fitted to the slave pcb would not be getting used without these connections.

I hope this helps

 

[SSLtech]

[quote author="Rob Flinn"][quote author="khstudio"]Keith said that C & D should connect from Master to Slave PCB... Not to the Slave switch PCB.[/quote]

Points "C" & "D" are still points "C" & "D" whether you make the link on the switch pcb or the main pcb, since they are actually connection points between the master & switch pcb's. i.e they exist on both master & switch pcb.
[/quote]
I think the typo is that it doesn't suggest interconnection between the main OR the slave PCBs, but instead talks about linking the two slave PCBs, when it should be talking about linking the two MAIN PCBs...

[quote author="khstudio"]...very strange that this "section" is needed twice :Shocked:

You wouldn't think so lookin at the schematic & the way the path is going - with only one Mak-up & threahold control... but OK, I'll do it. [/quote]
I had the same thought, but after building it, (which was when the photo was taken) I tested it out and then made the revisions which I listed in the updates. You can try it both ways also, but with the 072 and the other links I noticed I'd missed out is how it works best.

Either way, I think there should certainly be enough info now to make it all happen.

:thumb:

Keith

 

[Rob Flinn]

[quote author="SSLtech"][quote author="Rob Flinn"][quote author="khstudio"]Keith said that C & D should connect from Master to Slave PCB... Not to the Slave switch PCB.[/quote]

Points "C" & "D" are still points "C" & "D" whether you make the link on the switch pcb or the main pcb, since they are actually connection points between the master & switch pcb's. i.e they exist on both master & switch pcb.
[/quote]
I think the typo is that it doesn't suggest interconnection between the main OR the slave PCBs, but instead talks about linking the two slave PCBs, when it should be talking about linking the two MAIN PCBs...

[/quote]

It does talk about linking points C & D :

3. Link the following points

1.Point ‘C’ on both boards (output of detector, after the diode).
2.Point ‘D’ on both boards

Where it says, & I quote

9. Connect “C” & “D” on the slave main pcb to “C” & “D” on the slave switch pcb.

seems to be where the confusion has arisen, & is necessary otherwise the ratio contorls for the slave board would not work. Maybe I should word it "Connect the Slave motherboard to the Slave switch board via points "C" & "D". But I'm not really sure that this is clearer ?? What do you think ? It is easy to change it !

 

[khstudio]

Quote:
9. Connect “C” & “D” on the slave main pcb to “C” & “D” on the slave switch pcb.


seems to be where the confusion has arisen, & is necessary otherwise the ratio contorls for the slave board would not work. Maybe I should word it "Connect the Slave motherboard to the Slave switch board via points "C" & "D". But I'm not really sure that this is clearer ?? What do you think ? It is easy to change it !

This is definitly where the confusion is... Please look at the slave "Switch" PCB! C & D - DO NOT CONNECT to anything so that part does not make sense.


#2 -

Connecting the boards together.
5.Point ‘A’ on both boards

This is another confusing one... if point "A" is connected on the MAIN boards (not switch PCBs) I think it will connect the Detectors TOO early.
NO? :?
I could not find where Keith mentioned that one... Please correct me if I'm wrong. This "should" be more clear.

Also "B" should be added into the instructions (along with "A") -
It should read: "Connect A & B on each Main PCB to their corresponding "Swich PCB" - RATIO "Switch" control connection point.

:idea: I think you are confusing "C & D" with "A & B" :idea:


Kevin

 

[Rob Flinn]

[quote author="khstudio"]

Quote:
9. Connect “C” & “D” on the slave main pcb to “C” & “D” on the slave switch pcb.


seems to be where the confusion has arisen, & is necessary otherwise the ratio contorls for the slave board would not work. Maybe I should word it "Connect the Slave motherboard to the Slave switch board via points "C" & "D". But I'm not really sure that this is clearer ?? What do you think ? It is easy to change it !

This is definitly where the confusion is... Please look at the slave "Switch" PCB! C & D - DO NOT CONNECT to anything so that part does not make sense.


#2 -

Connecting the boards together.
5.Point ‘A’ on both boards

This is another confusing one... if point "A" is connected on the MAIN boards (not switch PCBs) I think it will connect the Detectors TOO early.
NO? :?
I could not find where Keith mentioned that one... Please correct me if I'm wrong. This "should" be more clear.

Also "B" should be added into the instructions (along with "A") -
It should read: "Connect A & B on each Main PCB to their corresponding "Swich PCB" - RATIO "Switch" control connection point.

:idea: I think you are confusing "C & D" with "A & B" :idea:


Kevin[/quote]

Kevin

Right I got you now, I think you are right about linking points A as well I have trawled back throught the thread trying to find where that came from ????? I will post the info again & remove the original, because I hate having to look through a whole thread for amendments. Kevin can you check them & let me know whether they seem to add up ?

 

[Rob Flinn]

Ultimate GSSL mods

You need 2 sets of boards (2 switch boards & 2 main boards).

Main pcb’s

Build the 2 main boards complete with the following exceptions:

1. Omit 1 of the 47K side chain mix resistors on each board.
2. Link Left + input to right – input on both boards
3. Link right + input to left – input on both boards
4. Do steps 2 & 3 above to the outputs of both boards.
5. Change the 10R resistors that feed the 12v regulators to 300R (This is when using an 18v mains transformer). This helps the 12v regs to run cooler.
6. On the slave board omit the 3M3 & 2K resistors next to where the board connects to the switch pcb.

Switch pcb

Build the master switch pcb up as normal, but make the following changes to the slave switch pcb:

1. Omit all resistors & caps that are to the right of where the word meter is printed on the board. This should leave only the 7 ratio resistors & diode.

Connecting the boards together.

1. Connect the master main & switch pcb normally
2. Connect the meter & on l.e.d to the master switch pcb
3. Link the following points

1.Point ‘C’ on both boards (output of detector, after the diode).
2.Point ‘D’ on both boards
3.Point ‘E’ on both boards
4.Point ‘F’ on both boards
5.The +12v & -12v points on both switch pcb.
6.Link both grounds, if they’re not already linked through the common PSU 0v, or elsewhere.
7.Link both pcb’s VCA CV (at the jumper link next to the 1K resistors in front of the VCA marked left that feed the CV to both VCA’s is convenient).
8. Connect A & B on the Slave Main PCB to the slave Switch PCB - this is the RATIO "Switch" control connection point for each board.

On the master switch pcb use 2 poles of the ratio switches (it’s normally 4 pole). One links to the master pcb as usual. Connect wires from the slave switch pcb to a 2nd set of poles on the switch.

If you intend to defeat the make up gain in bypass then link the 2 points where the defeat switch connects to then connect the switch to the master switch pcb in the normal way.

Other mods.

To have the option of Gyraf style detection lift one leg of the 47K resistor on the slave main pcb, on the side that feeds the 100pF capacitor and pin 1 of the side chain VCA. Connect it with a piece of wire to the pole of a spdt switch. Connect one throw of the switch to feed the slave side chain (where you lifted the 47K from), which is Oxford mode. Solder the other throw of the switch to the summing point on the master main pcb (the hole next to the 47K summing resistor where it feeds the detector). This is Gyraf mode.

For unity gain swap the 15K resistor next to each 5532 with a 15k trim pot. Center-travel should give you approximately unity gain, adjust on final test.
To reduce the threshold sensitivity, increase the value of the 47K resistors that feed the side chain, maybe double them or larger.


Trimming the VCA`s

When trimming the VCA’s omit one 5532 from each main board & trim the appropriate VCA. Then pull the 5532, insert it in the other socket & trim the other VCA. Once both VCA’s are trimmed you can insert both 5532 & you compressor is ready for action!

 

[khstudio]

It looks pretty good. :thumb:

Everything you wrote up was on time except the things I mentioned.

The way you have it worded still makes it unclear :!:

8. Connect “A” & “B” on the slave main pcb to “A” & “B” on the slave switch pcb.

The word "EACH" is important... you only mention the "Slaves".
If I was reading that for the first time I would NOT get it... I'm just trying to help & save others the headache I went thru for the last few days.

I think this is more clear:
"Connect A & B on each Main PCB to their corresponding "Swich PCB" - this is the RATIO "Switch" control connection point for each board.

Also, I spoke with Keith again about WHY he added the other TLO72 (+ extra jumpers) on the slave board & he said it had to do with the "Make-up gain" :!:
I looked at the schematic for a long time & didn't understand why that section was needed twice until he said that... it all made sense after he explained it... so that may be a good NOTE to add to the end of the instructions for others who are picking apart how & why things are done on this version.

Thanks Rob,
Kevin

 

[Rob Flinn]

Kevin

I have changed the wording but adjusted yours because it already says to connect the master main & switch pcbs as normal & I thought it would possibly cause confusion to ask people to do this again for points a & b.
Anyhow see if your happy with it & let me know.

 

[SSLtech]

Oh key, dough key.

I have no probe with re-wording for clarity. I'm a terrible offender when it comes to writing stuff that nobody else can understand! :wink:

Keef

 

[khstudio]

You know I thought about the wording last night... I'm sorry man, your right... it is already clear that the main should connect to it's swich PCB as normal. I think it was just the way I was looking at it, trying to picture what's going on... no big deal. I'm about a few hours from firing this thing up... can't wait :green:

Thank Rob & Keef (is it Keef or Keith :?: )

Kevin

 

[khstudio]

BTW,

I had a long talk with the main guy from:
IMS Distributor of Meters, Instruments
http://www.metersales.com

For the guys in the US that are looking for the real meters, he may be starting to carry them & keep in stock after my conversation with him :green:

What I'd like to know is, there are a few different models that seem like they'll work but just to be sure - which ones (Sifams that is) i did read all the other threads but they still left me wondering.

I know the correct one is the Detector 14 PPM 0-1ma.
I have read about the AL19 & AL19b, etc... & some other alternatives.

The question is, which one(s) will work correctly & is there a cheaper alternative from Sifam? The guy is waiting to hear back from me & I'd like to give him the correct info.

Thanks,
Kevin

 

[khstudio]

Important Question :!:

Chassis Ground? Should there be a connection from the Audio GND to the chassis at any point & where?

On the 1st & 2nd units I built I used regular 1/4 in jacks that grounded the sleeve to the chassis but this time I'm using XLR's which "float".
What should I do?

I also wonder if my first 2 units with grounded sleeves are not right or if it could effect the performance in any way?
If so... I would like to fix them ASAP.

Thanks,
Kevin

 

[Rob Flinn]

I connect mains earth to the chassis near the input sockets. This is my star earth point.

I then, for the "ultimate" connect from this point to both the input xlr's.

I also connect the centre tap of the mains transformer to this star earth point, so that it connects to the boards via the i/p xlr's.

You may be able to see what's happening in this picture.

 

[SSLtech]

Ground:

Every metal case needs to be grounded to the incoming electrical safety ground. That's just basic safety.

It is NOT always a good idea to connect the 0Volts to the chassis, nor pin 1 on the incoming & outgoing XLR sockets. That's how ground loops start.

Of course, for a piece of equipment sitting alone and unconnected to anything else AC-powered, then connecting the two makes sense. But when two devices built in such a way are connected, you get a ground loop; simple as that.

The very BEST way is to wire the inside with TWO grounds: an electrical 'safety' ground, -which must be the chassis- and an 'audio' ground (sometimes called "technical ground"), which is the 0 volts. then wire a switch to link or unlink the two, just like a ground lift on a DI.

Habitually linking audio ground to electrical ground inside every piece of gear is bad practice, and causes ground loops. The noise from the ground loop is rependant on the loop area, which means that many times everything contributes a little bit of the total hum, making silencing installations almost impossible.

In the case (no pun intented) of Rob's installation above, I'd run the two wires from the XLR pin 1 grounds along with the transformer 0 Volts, and link them at one contact of a switch, and I'd then run the OTHER contact of the switch to the chassis safety ground post. The switch can be on the back panel... it gives me a quick and easy way to make the unit compatible with ANY installation grounding philosophy.

Keith

 

[Rob Flinn]

Keith

What you say is very true, but making this connection for the GSSL doesn't cause problems "in my system" in original or Ultimate variations.

On another note I do get some hum in my system if I really go looking for it & was wondering what a typical level below zero ?

 

[SSLtech]

...It depends entirely on the loop size & area, so you can't really put a number on it. It also only tends to be noticed when it meets other tiny hums and "gangs up". It can also be usually cured by lifting pin 1.

I tend to keep grounds seperate in all gear -even purchased gear- then tie all the grounds to safety earth -or a technical earth- at the patchbay. That way I have REAL ULTIMATE POWER over hums.

Inky black silence.

Keef

 

[khstudio]

DONE :green:

I have to mix down a band tonite so I'll be testing it out but it does work pretty good & my first impression is it tracks better on my stereo buss.
I DO NOT like the Benringer meter though I want a real one that responds correctly!!! What are my options? It doesn't have to be exactly the original, just right. :wink:

Still have to adjust distortion trim & meter.
I'm also not used to hearing these 202X's but they are cool... a little colored/dirty but it could be the trimmers. I cranked the threashold & ran some stuff thru & adjusted by ear & it cleaned up a hair but it's VERY hard to hear like that, subtle but helped.

Thanks for your help,
Kevin

 

[Sorr]

Keef,

"I tend to keep grounds seperate in all gear -even purchased gear- then tie all the grounds to safety earth -or a technical earth- at the patchbay. That way I have REAL ULTIMATE POWER over hums. "

Do you mean you disconnect the audio ground from the chassis ground of each unit and connect it to one common earth at the patchbay?