Official Forssell Limiter Build thread

[clintrubber]

Sounds like some weird messing up of signals/cancellations/whatever. Since the actual gain-reduction is this simple, how about trying this:


- Use only the sidechain & the gain-reduction section (= two resistors & the Vactrol IIRIC).
- Drive both the sidechain-input & the gain-reduction section from a known good source with adequate output muscle.
- Probe the signal at the output of the gain-reduction section.

This should work.

If it doesn't then there's something sick in the sidechain (try various pot-settings) or you've been able to develop an error in the three-device gain-reduction section. How sensitive are those Vactrols to soldering heat ? Have you inserted them OK ? Have you inserted them OK ? Have you inserted them OK ? An error here will less easily show up in a DC-voltage test.

Then add the gain-make-up, if it's OK then the input etc.

Good luck,

Peter

 

[Svart]

do you have BOTH the DOA and IC options populated? you should have either but not both. the instructions are not clear on this though. It's the only way I could see that you are passing any signal without the DOA.

 

[babyhead]

Passing signal through the feedback loop?

 

[Lowfreq]

Hey Guys, Cheers for the suggestions.

Unfortunatley, I'm at work so I cant work on anything for at least the next six hours :evil: , but I have been looking over the input buffer schem, the parts list and the board layouts, and I think I might be onto something.... :?

From memory, I stuffed the board according to the parts list, and only really gave the schem a quick look over. Now on the parts list C33 is not included for the DOA version (only the IC version), but after looking at this,
http://www.diyfactory.com/projects/forsselllimiter/input_buffer.jpg
it looks like C33 is used in both the DOA version and the IC version.
I had a closer look at the board layout also, and it definitley has C33 connected to the in+.
Without C33 it looks to me like I'm connecting in+ directly to ground via R30.

I can see and understand why C32 isn't needed, and I haven't put that in, but C33 dosen't look like it should be excluded.

I'll try this at home, but I thought I'd ask if those who went the DOA way used C33? I hate not being able physically try this, but I guess i can just do the best I can here. :wink:

Am i correct in thinking that my buzz issue is related to this cap, as it's connected to ground alongside R30?

Am I way off?? It's a bit hard to read through the whole forrsell thread without my boss kicking my arse back to working. ha ha

Anyway, I'll try when I get home and also try clintrubbers suggestions with the sidechain. Now I just need time to go a little faster so i can go work on this baby. :green:

 

[kooma]

[quote author="Lowfreq"]..Sounds like a gate or a bad connection (I'm guessing bad connection)
Yet when I take the input DOA out, the compressor passes signal. :shock: :?

[/quote]
when I was testing my opto project I got same results, when opto was connected wrong. ie.->sound like pr1 is in series with signal?

 

[hejsan]

[quote author="kooma"]when I was testing my opto project I got same results, when opto was connected wrong. ie.->sound like pr1 is in series with signal? [/quote]
Maybe you have the upside down version of the PCB?

 

[Lowfreq]

Well, I've had a bit of a mess around, and checked a few things, done a few tests, and seem to have come up with a working solution. :shock:

My sidechain seems all good from what I've tested, so I set my input buffer with the extra parts for the 5534 IC to see how that would go, and my comp seems to work sweet as now. :green: :thumb: :thumb: :green:

The DOA input still shuts down like a gate, and I swear I've checked those connections 100 times, but the comp sounds good even with the IC input buffer, so I might just leave it at that for now, as I've got some mics waiting to be finished..........I'll come back to it, once they're done.

I did have to drill out the holes for the DOA pins a bit, so maybe I've got a bad connection there, due to not much of a solder pad being left.
Anyway, onwards to the mics. :grin:

[quote author="babyhead"]Passing signal through the feedback loop?[/quote]After looking at this part of the cuircut a bit, that's that only thing I could see that would let the signal pass through.
Even though I've built a few things now, I'm def still in the newbie camp when it really comes to understanding my curcuits, although everytime something goes wrong with any of my projects I seem to gain a lot more knowledge than when things work first time.

[quote author="khstudio"]I feel that opto type compression (at least the La-2A & possibly like my Avalon 747 OPTO - which uses a Vactrol too) are very musical. They seem to wrap around the sound & while controlling it, they seem to maintain a lot of character & life... and attack.

Different animals indeed.

Opto, FET, VCA[/quote]
I noticed with my quick try outs of this comp on voice, that some faster transients seem to sneak through a bit, in comparsion to my G1176, where nothing ever sneaks through. But I'm guessing that's more just the nature of opto's in general before I start trying to make my attack faster.
I wasn't singing either, just the classic "Check 1 2" :roll:, so maybe it wasn't the best test for it.

Is that why the 1176---->LA2A combo is used so often? 1176 catches the peaks while the LA2A does more of an overall smoothing??

Even my non-audio friend noticed the different flavour of this comp compared to my G1176. Her Comment "your voice sounds different" which to us geeks means a different flavour. :wink:

To finish this unit off, I'll do a bit more reading through this thread about adding an output pot to the DOA output, and how to set up VR2 for PR1 & PR2, and also get my meter at the correct scale. Then I'll be done!!

Thanks for all the help and suggestions guys!! :thumb: :thumb:

 

[khstudio]

I did have to drill out the holes for the DOA pins a bit, so maybe I've got a bad connection there, due to not much of a solder pad being left.

Dude, if the holes you drilled are "THRU" holes to the other side you may have drilled off the thru-ness :idea:

Check it out.

Kevin

 

[clintrubber]

[quote author="Lowfreq"]The DOA input still shuts down like a gate, and I swear I've checked those connections 100 times, but the comp sounds good even with the IC input buffer, so I might just leave it at that for now, as I've got some mics waiting to be finished..........I'll come back to it, once they're done.[/quote]

Is that DOA itself without suspicion ? Dunno what you're using there.

 

[Lowfreq]

I've tried different DOA's, and the same thing happens, so I'm def thinking I've prob done something when drilling.

It has to be in+, in- or the output pin on the DOA sockets, as V+ and V- are giving me good voltage when measured from Com.
When I get a chance I'll make some connections via wire, from those pins to wherever thy're meant to be connected to in the curcuit to see what happens.

Anyway, I'm quite happy now I've got this thing up and running. :grin:
It's functional and sounds good, which is the main thing, as I've got some recordings to do this week.

I'll finish all the fiddly stuff, once the recordings are done.

 

[joe-electro]

[quote author="Lowfreq"]I've tried different DOA's, and the same thing happens, so I'm def thinking I've prob done something when drilling.

It has to be in+, in- or the output pin on the DOA sockets, as V+ and V- are giving me good voltage when measured from Com.
When I get a chance I'll make some connections via wire, from those pins to wherever thy're meant to be connected to in the curcuit to see what happens.

Anyway, I'm quite happy now I've got this thing up and running. :grin:
It's functional and sounds good, which is the main thing, as I've got some recordings to do this week.

I'll finish all the fiddly stuff, once the recordings are done.[/quote]


Sorry, I'm not familiar with the schematic but the problem you're describing sounds like 1/2 of a balanced circuit is not getting signal. I've had exactly the same thing happen when a mic cable feeding an op-amp running in differential configuration has one conductor open. Hope this helps...

 

[tomcat]

Post deleted. Found all answers after another read through this thread :grin:

 

[tomcat]

Btw. imho i found an error in the modified 5534 output shematic (for the gain pot):

http://www3.sympatico.ca/tommytones/PCB_Stuff/Net_Pics/5534_Out_Changes.jpg

In this shem C39 is labeled wrong. It should be C38.

---

Btw. even this shematic is different then the 1C (mirrored) revision i have here... Still try to figure out.
Look at the gainup pot and hows it is connected to the opamp and R42.

On the PCB:

Seems like the output of the 5534 is connected to the gain pot only.

The wiper and other end of the gainpot is connected to R42, C38 an R39. So if you bridge the gain pot its ok.
If you use the gain pot all output runs over the gain pot, contrary to the shem above.

@Tommy: maybe you can post the silkscreen and layers of the 1C boards? I found only the 1b versions.

 

[Black Dog]

I made the following changes to the PCB Rev 1.C from Gustav:

Cut the trace between R42 and C38(C39 in the '5534_Out_Changes' schematic) coming from the output of U1.
Added a jumper from the output of U1 to C38.
Replaced R42 with a jumper to achieve unity gain(or thereabout - close enough anyway).

I'm using TL072 in the sidechain and replaced R24 and R25 with 10K resistors as Clintrubber suggested.

It sounds nice :thumb:

Cheers :guinness:

Lars

 

[khstudio]

Deleted

 

[khstudio]

I don't know much about SERVO's & want to use these Yamaha Discrete Op-Amps I have for the I/O buffers (NE80200).

Besides adjusting the feedback cap, do you think the servo will work with these Op-amps? They are the 990 footprint, can run off 24v & sound very good. I've used them for an API style pre & they work great.


Also, was the THRESHOLD issue of being TOO sensitive ever resolved?
Was it just the LIN. POT instead of the LOG that was causing the problem?

When using the DISCRETE Output (on Rev 1.C PCBs) do I still need to cut the trace to add the make-up gain pot?

Thanks.

 

[khstudio]

Also, does anyone have an overlay for the PCB they could share.
I've seen the one Frank posted but I could use the one with parts #'s for REV 1.C.

I know this project is older & I hope it didn't die.
Hopfully there are still some heads around here that can help out :roll:

TOMMYTONES?

 

[JonasWagner]

Hey Kev,

I got an overlay pdf for Rev 1b. Does this help? if, send pm.

Chris

 

[khstudio]

OK,
I finally got my ass moving on this project & I'm 80% done. :green:

Fred said that a JFET Buffer will work the best for the output bufer/amp so...

I'm using these Yamaha Discretes I have & don't think they're JFET input... could I (or SHOULD I) use the JFET 2n5457 stage (originally used for the IC output) in FRONT of my Yamaha DOA output amp?
(This is assuming the 2n5457's can handle the 24v rails!!!)

And if so... should I keep the 100k to GND before the amp? (after the JFET 2n5457 stage) OR should I go with the 10k... like the IC stage has.

I still haven't tackled the output gain issue yet but after reading I found that making the 100k to GND a pot should work :roll: Using the wiper to the + input of the output DOA.

Lastly,
What type of caps should the .33uf & .68uf cap be :?:
I'm guessing films?

Also, do they HAVE to be those exact values?
I have .47 & 1uf's in stock.

The .33uf is for the input of the sidechain & the .68uf is for the SERVO... will upping these values slightly mess things up?
I have NO experience with SERVOS.

PLEASE HELP ME OUT.

 

[khstudio]

[quote author="buttachunk"][quote author="khstudio"]could I (or SHOULD I) use the JFET 2n5457 stage (originally used for the IC output) in FRONT of my Yamaha DOA output amp?(This is assuming the 2n5457's can handle the 24v rails!!!)[/b][/quote]

the simplest is to use one output buffer or the other. but, you can tie the gain pot from my circuit into Fred's buffer amp by; sending RV1 from my buffer to J1 of the Forssell buffer. Don't change anything else. :thumb:[/quote]

Thanks for helping :thumb:

But i'm not fully understanding what you mean

Fred suggested "HIS" DOA with a Jfet Input... I don't THINK mine has a JFET input.

So I was thinking about cutting a few traces & using the 2n5457 SECTION ONLY, then going from RV1 (10k POT wiper) into my Yamaha DOA for the output amp.

Are you saying this is OK & to use the 10k for my gain control :?:

I've been doing a lot of research on this comp & really want to go 24v with the 2604's & my DOA's (they can handle it + I have exactly 4 spares)

Thanks man :thumb: