Official Forssell Limiter Build thread

[okgb]

Hey Kev , you gotta call the kettle black sometimes,
and i appreciate you having an opinion , the fact that
they are detailed lets me know that you've tried things
and not regurgitating what someone else has said .

So as i never have enough time for honest d.i.y.ing
getting a head start from someone else's efforts is a bonus ,
thank you .

[ the Forssell is 10 down on my list but i have the boards ,
got the t-shirt , gonna do that ]

 

[khstudio]

[quote author="Greg"][quote author="khstudio"]I'm sharing everything I'm going thru so others don't have to go thru this much BS.[/quote]
It's not BS, it's called circuit design. Ultimately the time and effort you've spent will make your compressor that much more special and unique.

[/quote]

Greg,
You are right. I learned a **** load from all the studying I've done on this thing. I also have a better understanding of the circuit, which in turn enables me to help others.

The BS I'm referring to is other problems & missing or scattered info... like not having an up to date overlay of the CURRENT PCB, the Threshold being wrong on the PCB, phase being reverse, the bridge rectifier not being on the PCB or on any schematic (except Franks Drawing - THANKS FRANK :thumb: )
ETC...

I will say I'm VERY thankful for all the hard work others have put into this & the info they DID share... this IS a GREAT COMP. :thumb:


I'm getting ready to try the second half of the SERVOs 2604 for the output buffer.

 

[dissonantstring]

good job kevin! :thumb:
i just read this thread and haven't even looked at my forssell boards in about a year, but i'm sure when i do get to them i'll be using all of your advice.
thanks for all your hard work. i must say that when you go, you GO!
i'm amazed at how much you accomplish in such a short while, like with the Night EQ. kind of infectious i must say.
cheers,
grant

 

[khstudio]

Thanks :green:

I'm really NO Master tech... by no means.
I just do a LOT of reading, studying & note taking.
Not to mention bugging the **** out of some people to learn :razz: ... but I ALWAYS share what I learned.

I knew nothing about electronics about 3 years ago & have come a LONG way... the hard way. :wink:


I got my DOA pulled (had to solder the Yamaha DOA's... damn square pins don't fit the holes)

& I'm about to hear the 2604 for the output.
Be right back with the news.

 

[rattleyour]

Hey Kevin,

I just stuffed several 312 cards with Yammy DOAs-- used a small file to tame the square edges, worked really well-- although they are still very snug.

Might work for you--

Liam

 

[khstudio]

It works :green:

The 2604 for the output buffer sounds Very good... not as much body & depth as the Yamaha DOA but a little cleaner/FASTER! The lows/SUBs are very tight & Less full... but nice. The Yamaha has killer low end & nice mids + highs are smooth, less there but more open somehow - compared to the 2604.

I even tried going directly in (unbalanced) from my NEVE 1272 (Tranny Balanced) & THAT sounded awesome... I just shorted pins 1 & 3 coming into the FF Comp. So for everyone with Tranny out pres, etc... you don't HAVE to use an input tranny or amp for differential input... not sure how it would work otherwise :?

I'm sorry I'm focusing more on the buffers than the comp itself... but what can I say, the sidechain (once built correctly) is KILLER :thumb:

Just remember... when your in bypass, just running thru the comp, that raw signal is what's going to be compressed. So if you don't like the sound in BYPASS, ypu won't when compressing. The AMPs, Trannys & components you choose will determine the "FLAVOR" of this thing.
IMO the sidechain has NO "Sound".

I have decided on my build to NOT use any input buffers but I'm undecided on the output... they're all cool for something Options, Options.

Now deciding weather or not I should get Freds J-990-3 :roll: .

Greg,
I'd like to take you up on listening to the amp but I don't have any pin/sockets & DO NOT want to solder your pins :!:

 

[dissonantstring]

hi kevin,
just to clarify, when you say you are bypassing the input buffer you are essentially taking the pin 2 input from the XLR input connector and directly wiring it to the 4K99 resistor and tying pin 1 and 3 together (unbalanced)?
is that correct?

also, when you are using the 10K as the output control did you mean the resistor noted as RV1? did you use a makeup gain pot as well?

this is pretty exciting. your experiments with Fred's guidance has cut down the parts count. great job!
-grant

 

[khstudio]

[quote author="dissonantstring"]hi kevin,
just to clarify, when you say you are bypassing the input buffer you are essentially taking the pin 2 input from the XLR input connector and directly wiring it to the 4K99 resistor and tying pin 1 and 3 together (unbalanced)?
is that correct?

also, when you are using the 10K as the output control did you mean the resistor noted as RV1? did you use a makeup gain pot as well?

this is pretty exciting. your experiments with Fred's guidance has cut down the parts count. great job!
-grant[/quote]

Pin2 + to the 4.99k
Pin3 (& 1 optional - depending on ground issues or not) goes to the circuits ground. I linked mine without any problems.

Fred said you should put a small electro cap in line just to be safe (from DC) on the + input to the 4.99k... 10 to 20uf max is what he suggested.
He mentioned liking Nichicon PW's in the signal path (just like me :grin: )
He also likes MUSE KZ's.

Look at the DOA output schematic (this is what I'm using & reffering to) the 100k from + to ground BEFORE the DOA is where I used a pot instead: wiper to the + input.
Fred said I should lower this to 10k LOG

 

[dissonantstring]

thanks for the clarification kevin! :thumb:
-grant

 

[martthie_08]

I'm finally getting around to building this comp and have started to stuff the board with discrete input and output sections (DIY 990 opamps). The comp will be part of a discrete channel strip running on +/- 24V with an external PSU. So this is basically a "hello I'm in the FF opto boat" post, I still have lots of reading to do and will post my findings and progress here.



cheers, Marten

 

[khstudio]

:thumb:

The main thing you'll have to deal with is how to implement the Make-up gain when using DOA's... the output is rippin LOUD without it.

Good luck.

Still haven't boxed mine up yet but it's sounds great. :green:

 

[martthie_08]

It feels exciting to work on a project that is not so very popular in this little community. The fact that is not so well documented means there will be lots for me to learn, I am sure about that!

One quick question: Where can I find info on determining R28? I guess this resistor sets the maximum ratio, but how is it calculated for given ratio, say 8:1 :roll:

and yes, I'll have to look into implementing the variable makeup gain...

 

[khstudio]

I THINK it's 100 ohm. :?:

Every time I start a new project, I go thru the threads & copy & paste any & all good info into a word document or text file I can read later. (Over & over again :wink:

Like R28... you will find it mentioned about 5-10 times with pretty consistant answers.

I also print it out & highlight all the good stuff + write some of my own notes.
I would download ALL the schematics & info you can from this & all the other threads or sites (like Kevs site) concerning this project so you can get a grip on it all.

My recent posts have some good info too.

The comp will be part of a discrete channel strip

If this is the case I don't think you NEED or SHOULD USE the first DOA... it's for un-balancing the input. If you have a pre in front of the FFO it will be doing that already.

Please get a second opinion just to make sure.
:wink:

 

[martthie_08]

I THINK it's 100 ohm.

yes, thanks :thumb: I just found it in this thread, 100R is what everybody is using. I use Scrapbook in Firefox to make long threads available offline and to enable fast search and it turned out that I forgot to update the search cache, so nothing was found for "R28", but now it works like a charm. :grin:

I'll try what the input buffer does and will decide then whether to go unbalanced or not. The comp will sit between an API style preamp and a Siemens W295 EQ, I'd like the individual modules to be individually accessible for inserts...

 

[khstudio]

I'd like the individual modules to be individually accessible for inserts...

That makes sense now. :green:

I use firefox too... I should try that out.

 

[BR]

Man... I have two of these boards sitting here and I've been dying to try them out. I still haven't got around to ordering the parts.
Do you have a BOM for this Kevin??
Thanks for all your trouble figuring this out. I'm getting excited about this project yet again.
thanks
Gil

 

[louder]

Hi
the problem that some are talking about,info not focused,in my opinion,it`s because there are at least two boards,and several ways to put them to work.
i got two red boards from tommytones.
others have boards from gustav,which i think are somehow diferent.
what i think we should do,it`s that one always should mention which board rev one is talking about.
seems to me that the info it`s been given,but it`s alloer the several posts.
i did read lots of times,the threat about this comp,and every time i get to the end,i fell uncertain.
those who already got there comps to work,please step forward,cause i fell there`s lots of guys like me.
sorry for taking so long
best regards
pedro

 

[Nele]

I built two of these units, one with DOA's and one with dip's and had no problems whatsoever.. fairly straightforward and simple build.
Everything you need is right here.

Definitely worth building!

Cheers,
Corneel

 

[martthie_08]

so, more progress on my build of the FF opto: Pots and LED metering in place



Kevin thanks again for your advice, I started by leaving the input buffer out as I don't have the DOA's ready yet (still need a damn LCR multimeter), but this gave me the idea to set up a quick test for the sidechain / metering. Even if there's nothing to be heard yet, the following movie clip gives me a good feeling :grin:

movie

now on to the discrete output stage and the makeup gain adventure.

cheers, Marten

 

[martthie_08]

ok, it is alive! The compressor got a cap at the (unbalanced) input and a 990 DOA / old Stancor 600/600 iron on the output (for balanced output) and what can I say - it rocks!




will the input cap (10uF/35V at the moment) need a resistor to ground and should I bypass it with a film cap?

the "threshold" pot is actually a "compression" pot - correct? Now it is set up so that the pot is turned clockwise for more gain reduction. I did a test, the other way around all the action happens on maybe 2 degrees of the pot, not good.
[EDIT] This seems to be normal, I just found Kevin's remarks on this! [/EDIT]

the three OPA-2604 opamps get pretty hot, especially the one for the DC-Servo output stage - is this normal or should I scope it?

Next thing to do is check all the levels and calibrate the meter and adjust the opto matching, then add output control (makeup gain) and check the frequency response.

[EDIT]
the unit has a -4dB drop when used in bypass in my configuration (R6@10k, instead of 100k)

I couldn't find anything on matching the opto cells (with VR2) how does it work and how does it interact with the LED meter adjust (VR1)?
[/EDIT]

Kevin, what did you do to add gain control of the discrete output stage? I have read some of the previous posts on this and I'm not sure I understand correctly - replace R6 with a 10k log pot like this?