553 / Harrison Ford Filter mashup in 500 series; thoughts?

I've been toying with this idea for a bit, as I'm currently sitting on a few stuffed Harrison Ford Filter pcb's and wanted a way to use them in my 500 series rack without making them an entire module.  I also have a few 10k linear center detent pots sitting around, a couple of 2503's, and recently have started winding some inductors on rm8 cores.  So, next logical step is to try and pair the Harrison Ford Filter board with the API 553 eq circuit (plenty of schems out there, one on Classic API or PeteC's 5003 schem floats about) in a 500 series chassis.

Here's a block diagram of what I've come up with (mocked up in the always useful Microsoft Paint), details below.



I decided to use "building block" circuits that others much more knowledgeable have made available.  I wanted the option to have a full bypass, so line in first hits the CAPI plug-in relay (http://classicapi.com/catalog/product_info.php?cPath=102&products_id=195).  This switches the output between the line in and EA2503 out.

Line in then goes to the Harrison Ford Filter PCB.  This PCB offers provisions for balanced line in and out, but I'm only going to use the balanced line in to a THAT 1246, taking an unbalanced feed after the filters.  These two filters can be switched in and out with a pushbutton switch.

Unbalanced out from Harrison Ford Filter PCB goes to a CAPI discrete voltage follower (http://classicapi.com/catalog/product_info.php?cPath=71_72&products_id=221) to act as a buffer between the two circuits.  The unbalanced out from the discrete voltage follower feeds the 553 circuit, which is currently a direct copy of the schematics.  Out to a 2503 wired 1:3, whose output feeds the other side off the CAPI plug in relay.

I threw together a quick schematic in KiCad.  This would house the 553 circuit and the rest of the inner workings/connections, wired up to a 500 series edge connector PCB (http://classicapi.com/catalog/product_info.php?cPath=87_88&products_id=402).  Please note, most components are unlabeled, and I really have not double checked it at all.



Basically, this is a direct copy of the 553 schematic, with provisions for the "building blocks" added in.  For the relay power, I copied that little bit from the vp312 DI schematic publicly available on the CAPI site.  It seems to make sense.

My idea for general faceplate layout:



I'll post some more tomorrow on the module layout, but wanted to start getting some feedback on the general idea and how best to implement it. 

Another note: I have no idea what I'm doing, but I've gathered enough knowledge to conceptually piece things together.  The more input that people provide, the more I learn. 

Much appreciated,

Keith

Edited: added a bunch of links

Just looking at the schemo, couldn't you use a SPST to bypass the 553 bands? I want to say that is how the original is but can't quite remember.. But I'm no expert 

I like the idea of variable filters. I have a pair of 5003's but rarely use the lower band or the extra frequencies. 

Indecline said:

Just looking at the schemo, couldn't you use a SPST to bypass the 553 bands? I want to say that is how the original is but can't quite remember.. But I'm no expert 

I like the idea of variable filters. I have a pair of 5003's but rarely use the lower band or the extra frequencies.

Click to expand...

The original schem that I've found uses a 4pdt pushbutton.

http://www.classicapi.com/catalog/images/reference_docs/API-553-schematic.pdf

I agree on variable filters. I see the HFF/553 combo as a very useful tracking eq that can hold it's own during mixdown. From what I've read though, the lower band on the 553 is supposed to be pretty good. can you elaborate some on why you don't use it much?

I was taking a look at the JLM INX5 pcb schematic, and realized I could use much of it and simplify the pcb I would need to make for the 553 section of the eq.  There's a few of em making their way across the Pacific right now destined for my doorstep. 

The INX5 also offers the option of a 2520 line receiver or any number of ic options (I would be using a THAT 1246 if ic, super easy solution... It'll be fun to play around with and hear the differences vs doa) on the line in and an integrated relay bypass.  This keeps the pcb I need to design almost free of power, except for the discrete voltage follower.  I've only ever designed and had made a passive pan/bus pcb, so this should make the entry into powered pcb relatively simple.



My usual board house is on Chinese national holiday for the next week, so I have some time to play around with the pcb layout.  Right now I'm looking at a 2.7" x 2.5" pcb which should easily fit with the INX5, HFF pcb, and a 2503 in a 500 series.  I'll post the layout once I get it "done".

For inductors, I've wound up some rm8 bobbins on n48 400 core for the mid band (100mh) and n48 630 core for the low band (1H).  Bobbins are hand wound; it's a chore, but a labor of love.  All you have to do is wrap a wire around a plastic bobbin a whole bunch of times.  Just make a mark at every 100 turns to help you keep track, and go to town.  The mid was ~500 turns with 36ga  and the low was ~1,300 with 38ga.  My camera is being uncooperative, I'll post some inductor pics once I get the camera sorted.

I spent a little more time messing around with this, and threw together a quick reference breadboard to work out a layout.

Schematic:



I used a 2" x 3" breadboard.  I had to cut one of the sides down to 2.8" for the first inch to leave enough clearance for the Harrison Ford Filter pots.





Everything fits, it's a little tight, and I have about another inch or so rearward clearance between where this pcb would sit and the INX5 pcb, so for the actual PCB layout I extended the front to rear to 2.5".  Proto PCB layout shown below, ground plane not filled in.



I ordered a 5 piece proto run from pcbcart.com, they are supposed to arrive Tuesday.  Hopefully everything works out! 

In the meantime, please feel free to point out any flaws you see. 

Thanks,

Keith

Making some progress.

The proto boards arrived from pcbcart.com, everything looks fine (except for the garbage picture, the lighting makes it look like a pretty poorly made board... but I didn't feel like taking another one).



And fully populated:



Front showing 4PDT switch location:



For the inductors, I wound up some rm8 cores.  For the 1H low frequency inductor, I used N48 630, ~1,300 turns.  For the 100mh mid band, I used N48 400, ~500 turns.

This is how everything will (in theory) fit nicely in a 500 series.  Note that the Harrison Ford Filter PCB will actually be upside down when all comes together.



It looks like it will be a bit too tight for comfort on overall width if I use the pushbutton switch on the Harrison Ford Filter PCB, so I decided to wire a mini DPDT toggle switch off board for this function.  Here's the overall fit once everything is populated.  (I haven't drilled the L-bracket yet, hence why it's backwards.)



From the looks of things, the pot and switch locations should provide a pretty usable face plate configuration.



Now I just need to finish wiring everything up and test it out.  Hopefully it sounds ok and nothing catches fire.

That's all for now, more to come soon!

Thanks,

Keith

Spent some time with the drill and a blank L bracket today and got everything wired up.









The unit I currently have working is about -6db on the output... after reviewing the Harrison Ford Filter schematic again, normally, the HFF circuit uses an INA134 (0db line receiver) and DRV134 (+6db line driver), so leaving the DRV134 out of the circuit (taking unbalanced) and using a -6db THAT 1246 would leave a -12db signal feeding the discrete voltage follower buffer, not the -6db I was expecting.  Putting a 1240 in it's place should keep the overall circuit at unity gain.

I'm just glad it works, and it sounds pretty damn good too.  I'll need to listen to it again once the euphoria over the fact that it actually worked subsides, though...  But so far, minus the line receiver issue above, everything is working exactly as expected!  Which is odd, because I honestly have no idea what I'm doing... 

Hey, great job!
Your idea was good to combine both, it's even better if it works 
I think your next step is to measure what's going on so that you can confirm the quality of the sonics?
Keep us tuned