About the input switches and stuff... my opinion is: of course we need phantom. polarity is nice when doing drums for eg. I don't like to pads input but if the input TX can't handle hi levels let use it. uhm... I just realize that we don't have output trim on API...
we may need pad!
so why not to have PAD/PHANTOM/FLIP ? 3 relays and 3 switches... no big deal. people can mount as desired.

James,

I don't think your R1.. 1k linear pot is right. I know I saw that on Dan's site but I think it is wrong. The data sheets he does NOT have on his site list the adjustable gain resisters from 0 to 37k in 6db increments. If you used a 1k pot you would vertually have no adjustment if I understand this right. I would think at a minimum you would want a 25k more likely as close to 37k as possible. Any thing above the 37k has no effect on the circuit.

I'm not too good with electronics yet but I'm trying like hell to learn.

Also Ulysses has told me that the input cap and resisters need to be modified to match the input transformer. There for he recomends a JT-110 with a 100k resister and a 31k plus a 56pf cap in series as replacements per the jensen site.

I'm confused if I really need to modify the input section. (I actually have a bunch of 325 cards that I'm converting over to 312's. It is just a change in caps and resisters plus the input tran)

If I use the stock input section of a 312 and use a jt-115 it does not match the spec of the tran. THe jt-115 uses only a 150k at the input.

Any thoughts?

IF someone has a scanner I can fax the missing sheets from Dan's site to them. But they would have to make them avaiable for everyone here,

Joe

WOW !! :eek:

I turn my back for a little while and look what happens !!
I actually went and had a sleep. ( I don't do that often)

So many very cool post and form so many cool people!

It's all good

I think all outstanding questions got answered.
You now now how things sound and how 1272 amd 312 differ.
We know people are looking at components in detail.
Bear has brought up the choice of sticking to traditional mounting ....or the NEW box approach. This could mean a slight rework of PCB to provide mounting and screw terminals etc. I think a pair in a single RU box makes for a great DIY, sensible and useful.

Keep it up guys ! I'll get out of the way know and just let you go.
Give it a couple of days and I think there will be enough info here to summerize this into a web site at Group DIY. Rmember James has a page with most of the raw data links so grab it while you can.

as I said, It's all so $%#&ing good

I will probably build mine in the traditional API format on a 3" card with a 15 pin connector.

The connector is about 4 bucks and makes for an easy removeal for repairs or mods.

Also they will fit the API 411 rack. I have bought two of these racks of e-bay in the last year for under fifty bucks each. Add a power supply and a i/o pigtail and your all set. A 411 rack can hold 11 325 cards or 8 312 cards. The connectors are mechanicaly moveable to accomidate either size card. (The 312 is about a half inch wider {or taller if laying down} with the tranny then the 325)

The pin config is different for a 500 series rack. A 312 card from Brent Averill that is set up for a 500 rack has some green wire mods to make it work. It is not real pretty.

At some point I may make up a few extra cards for later. These sound so damn good you'll want more. With a 411 rack you just keep adding cards with no extra work.

The other option I am playing with is to use a 2 space rack and imediately set it up for 8 cards on edge. Put all the i/o's and switches in now, and add cards later.

With extra cards laying around it is an inexpensive way to add channels. For a little over a 135 dollars a channel to build out you can't beat it.

Joe

Hi Joe,

Do you know where to get a blank 3" 15 fingered (is that the right term?) card?
I got 4 empty spaces waiting to be filled in my API 500 rack. I'm willing to build four in a new 1U rack, but would pefer to put them in my API rack.

Cheers,

elco

Originally posted by JRE Productions:
The other option I am playing with is to use a 2 space rack and imediately set it up for 8 cards on edge. Put all the i/o's and switches in now, and add cards later.

With extra cards laying around it is an inexpensive way to add channels. For a little over a 135 dollars a channel to build out you can't beat it.

Joe

This is the kind of layout I was thinking about. In fact it would be nice to have a 2 space rack with 4 Neve 1272's and 4 API 312's.

Is something like this possible?

Will these API 312's require the switch assemblies and the I/O cards or will the design be all inclusive?

Regards,
Steve

P.S. This place is incredible! What a tremendous resource it is and the talent of its members is awesome.

Originally posted by Flatpicker:
You can hear for yourself here at: Silonex Inc.

thanks for sharing this link.

Those API recordings sounded great. The Neve recordings sound sort of soft and silky, whereas the API's sounded crisp and clean.

Now I definitely think I would like to have a rack of about 8 of these API's, especially if the price runs about $135 per channel.

Thanks,
Steve

Steve, I'm going to build them as regular cards with no switches or controls atatched.

Off of the connectors I will add the switches and pot. Two reasons for this: I thinking of building an switch network on a small card that the switches mount to. THis can be used for any pre's you will make as a 20 db pad, a phase switch and a 48 v phantom will be common. SO once I build the first one I will just make a bunch more.

I guess every project can and will be different. Unless someone is making a PCB to fit my aplication, I will try to duplicate an original API card.

As far as 4 Neves and 4 API in the same rack? I usually would not mix and match. I think it would be better to do two seperate 1u boxes, one with API the other with Neve. (By the way the power supply is diferent for each.)

Even if you go with a standard API card layout with an edge connector you can get 4 in a 1 space box. Depending on how deep it is will depend on if yopu get a power supply in there too. THe original 312 cars is smaller than a 3x5 index card. Just a little bigger than a pack of smokes.

Elco,

I do not know where to get the blank cards yet but they are a standard size and many other products use this same rack and card size.

If I had a 500 series rack I would make it fit that size right off the back. It is the same 15 pin connector but it is a 5" card instead. Basically you could take the exact same PCB layout and use the 5x7? card instead. It still breaks out in the back to a 15 pin connector. Even with the extra space I would use the same smaller pattern for the traces. You would have to modify the design to change the pin out to match the 500 rack pins, and to insert the switches. Then I would make a face plate and mount your extra switches on it. Mount that to the card.

The other thing you could do is build a standard PCB non edge connecotor project like others here and then mount it as a daughter board to the 5x7 15 pin blank card and do the connections for switches and ins and out between the big card and the little one.

Joe

Here's what I'm thinking....

I initially wanted something similar to the Tim Ryan design. A single board with a connectors and switches mounted to the board that would also mount the board to the case. Seems like a pretty good idea to me. However based on the tons of variations here it seems that it would be best to break the boards up a bit. So now I'm left with this...

You're always gonna have the base board. This will contain the basic API circuit and output transformer.

The next item is the I/O board. There will most likely be a couple of variations of this board to accomodate different input transformers. This board will also have the relays for switching that polarity, phantom and pad.

Finally there will be the control board. This will contain the gain control, switches and indicators.

Roll Music already has a control board and an input board. Of these the control board is usable. The I/O board isn't really usable in my plan because it has no provision for transformers. I don't really care for the control board either because I'd rather have push-button switches than toggles. That's a personal thing though.

It's entirely possible to leverage the Roll Music I/O board if the main circuit board had provisions for the various input transformers. The four most popular that I can see will be the Lundahl 1538XL, Lundahl 1636 and the Reichenbach RE115k and Jensen JE115k. Of those I believe the Reichenbach and Jensen have the same footprint. The two Lundahl's do not share the same footprint. That's why I would lean towards moving the input transformer to the I/O board.

There's always the possibilty of making the API 312 project into a handful of "standard" formats. That way instead of bickering about which style to go for you gould have your choice. As always make your circuit boards available to the public and away we go.

Also if anyone has any information, pics, schematics etc that they would like for me to add to my API project page then let me know. Send it my way and I'll put it up. I'm particularly interested in the "missing" API pages. If I had a fax machine (and if faxes were truly legible) then I'd just have them faxed to me. Maybe I'll play with that fax/modem on the server....

Peace,

James

Originally posted by JRE Productions:
I am thinking of building an switch network on a small card that the switches mount to.

Even if you go with a standard API card layout with an edge connector you can get 4 in a 1 space box. Depending on how deep it is will depend on if yopu get a power supply in there too. THe original 312 cars is smaller than a 3x5 index card. Just a little bigger than a pack of smokes.

Joe

Are there already schematics available for the little switch network cards and the I/O cards or is this something you or someone will have to create from scratch? Will these little cards be created as PC boards or are they simple enough to just build from the schematic?

You say, that 4 of these cards should be able to fit inside a single 1RU case ... how might they be arranged? Do you suppose they will be laid out 2 by 2 with the edge connectors running down the center? With the cards moved all the way forward, will the case have enough room left for a power supply and the input transformers?

Sorry for so many questions ... if there is a sight or another message I can read that will answer some of these questions, please let me know.

Thanks,
Steve

Hi Steve,

Technically speaking I am less of an electronics engineer than a mechanical engineer.

As far as mounting cards, I would consider mounting them all facing forward with the connectors facing toward the back. Two reasons. First the would only take up the first five or six inches in a rack case. Second, it becomes some what of a nightmare to wire if you had them set up as 2 and 2 facing outwards, with the connectors inwards. This reverses the left side wires from the right side. So your power and i/o's would cross each other. I'm sure you could work it out if wanted to.

As far as switch cards, Rolls has them if you want to buy them. Also his new input boards have a spot for the input transformer right on the pcb. It is setup for a JT 110k though and the size is a little different than the 115k. He has had very good sucsess with the 110k trany in his API 3125 box. He just changed the input resister and cap to more appropriately match the 110k trans.

I will probably try to build my own. I have not made PCB's yet, but am gathering the parts to do so. The problem with buying them is it gets very expensive and can add another 100 per channel.

I'm sure I'll be asking the questions in the next week or so.

I do have a ton of API documentation and knowlede because of wanting to do this project for many years. I own the API 325 line cards (5)witch are basically the same as the 312 minus the input trany. I have mine connected with tranys for use as a pre. They sound freakn' great.

I will probably build 4 more. That makes 9. I think the 9th one will be a variation with a Hardy 990 and a JT-16 trany, racked by it self. A much more supper clean version, I suppose. THe 990 needs a little different input impedence than a 2520.

Joe

Hey Steve,

The stuff you're asking for (and would make the project a whole lot easier) is stuff that hasn't been finalized yet. I personally don't want to make edge card versions of the API. Obviously some people do. My take is that if you want the edge cards then make it compatible with the API style racks. The Rollmusic site has an API 325 to 312 conversion page. They also provide some building blocks that will help you out. I like the concepts of these building blocks and would like to change them around a bit to broaden the appeal of the the 312 project. Check out the Rollmusic project [[url=http://[/URL]="http://www.rollmusi…"]here.[/]="http://www.rollmusi…"]here.[/]

Their building blocks will greatly speed up the process of building an API project. The power supply should be strong enough to power several channels and the input module is good for 2 channels. To get four channels you'll need the power supply board, 2 input boards, 4 control boards and a power transformer. That'll bring the total for you up to around $370. The only thing that you'd have to do then is make the 312 card.

How you approach the 312 card at this point is entirely up to you. I'd like to see a couple of "standard" implementations surface. That'll make the DIY easier for everyone else. If a standard board layout pops up then maybe a group of people could pitch in on an ExpressPCB order. That way you won't have to etch and drill the PCB's.

I'm pretty stoked about the project. I'd like to see it develop into an easy to build DIY project. Time will tell.

Peace,

James

Yes I really like the modular concept as well. I checked out the Rolls site for the 3125 project and that looks exactly like the type of configuration I would like to attempt. The issue here regarding purchasing these modules from Rolls is the increased overall cost of approx. +$90 per API card.

Are there schematics available for these modules (or some like them) that could be incorporated into this project (or are we just not at that point yet)? They seem like they would be some fairly standard circuits; in fact the best of the available solutions could be used for each separate module.

I know there is some more work to do, but this project is such an awesome idea I wish I could do something to start ... yesterday. I am far being an expert in any of this, but if there is something I can do to help relieve the burden and possibly speed up the process, please let me know.

Thanks,
Steve

> There are notes on the API 312 schematic to replace C4 and C5 with "polystyrene" caps.

Polystyrene is so-1970's.

I remember when every PCB had at least one polystyrene. What a pain: soldering ruins them.

I think they were as-good-as-we-can-get when the alternatives were crappy ceramic or wax-paper.

For most polystyrene values, at today's parts/labor rates, I would look for the good Silver-Mica. Expensive, but very neutral (and stable like a rock). And will not melt.

For larger sizes than S-M has, use whatever audiophile film-cap you favor; Solen, BlackCat, GoldenArch, whatever. If too small for that, use a precision ceramic, monolythic if you can. Modern films and precision ceramics solve the problems we fought with polystyrene-caps.

It looks like you need 100pF and 220pF, better than 25V and 10%. DigiKey has C-D Silver-Mica 500V 5% in those values for $1 to $1.30 each. That may be the same price as in 1972; but back then $1.30 bought 2 packs of cigarettes or 4 gallons of gasoline. Today a buck is nothing.

If you are convinced there is a "Polystyrene sound" that this board needs, good luck. Even if you find a modern polysty cap, it may not be wound the same as the old ones. Try surplus cap shops and find some NOS caps of the right vintage.

[QUOTE...The four most popular that I can see will be the Lundahl 1538XL, Lundahl 1636 and the Reichenbach RE115k and Jensen JE115k...[/QB]

I never checked to see what type of input tranny the API used. I knew the API 2520 wasn't the same as a Jensen 990, but I always assumed it used the same 1:2 ratio tranny like the JT-16. I didn't realize it needed a 1:10. Has anyone ever calculated the voltage noise/current noise or optimum source impedance of the API? (explained here:
http://www.jensen-ransformers.com/mic_in.html)

I like the module idea so I'll be working this way: 1 board will be the amplifier (2520 + 312 devices). 1 board will be TX input transformer with all option we decide (jensen+lundahl+what else ?) with PAD/FLIP/PHANTOM and GAIN (pot) using 3 switches and 2 Aromat TQ2 relays. Someone would like to suggest other relay footprints?? The last board will be PSU with 317/337 regulators. I won't include option to edge connector since my goal is to rack 8 preamps in a 1U rack. If you have some suggestions let me know.

I will not be needing edge connectors. I like the idea of the modular design. What would be cool if the input board could be used with different preamps. It would contain pad/polarity/phantom/input tranny/gain. Anything else I missed?
A 325 card would be cool also. To use in a Pultech?
Jesse

Originally posted by Fabio Bauman:
I won't include option to edge connector since my goal is to rack 8 preamps in a 1U rack. If you have some suggestions let me know.

How do you plan on fitting 8 channels with power supply and the swith and I/O modules into a 1U rack?

Steve

I would vote for a common gain block arrangement (i.e. a 2520/990 footprint), with the compensation caps P/S decoupling caps on-board (as well as the feedback coupling cap), and connectors for input tranny, gain switch/pot, output tranny, P/S connections. This module should ideally be designed to be long rather than wide, as this would allow many to be sandwiched beside each other in a 1U case (it would just need to be deep). I figure that 3" X 1.25" would be more than enough..it could likely be even smaller.

Personally, I'd rather see things like transformers, switches, pots on seperate boards, as this lets folks decide their own chassis layout. Hard mounting XLRs and pots to main PCB's really limits your chassis choices (this was the one thing I didn't like about the 7th Circle Audio 1272 project).

I would also suggest a 317/337 based power supply, so that users can play around with higher voltages (if their op-amps can take it). Plus, it's been my experience that the 317 series provides better regulation than the 78XX series.

I'm looking forward to this project (started buying parts already).

Cheers,

Kris

Originally posted by Dr Frankencopter:
This module should ideally be designed to be long rather than wide, as this would allow many to be sandwiched beside each other in a 1U case (it would just need to be deep). I figure that 3" X 1.25" would be more than enough..it could likely be even smaller.

I'm looking forward to this project (started buying parts already).

Cheers,

Kris

With this type of configuration, how many channels do you suppose could be put into a 1U case? How deep would it have to be (or how deep can it realistically be)?

I too have begun acquiring parts for this project as well. It is going to be an awesome unit when finished!

Regards,
Steve

Well, I'd figure about 16" for chassis width (though I'd need to measure). Figure on the input transformer being about 2" long and 1" wide (if it's a Lundahl), or 1" diameter (Jensen). Just guessing, but figure on the output transformer being 2" square (we'll see when they arrive). I figure the gain switch will stick out about 1", as will the XLRs.

So, if the output trannys are 2" square, we'd need to staggar them in order to make them fit along with the power supply, so give a total of 4" depth for the output trannys. Okay, so for a stab at the required chassis depth, lets add up all these guesses:

1" XLR
4" OP Xfmr
3" Circuit card
2" Input XFMR
1" Gain Switch

For a total of 11".

Cheers,

Kris

I found that at the National Semi site. Looks like you could dial in the precise value you wanted up to 20V although the data sheet states that you lose some current at the extremes. Also, the +/- 25V (on the supply side) gives you a 50V line as well which could be locked into +48V for the phantom.

I'm sure some stuff will have to be changed a bit. For one those caps onthe output look small. I would think that something closer to 1000uF would be more appropriate. Not that I REALLY know what I'm doing there.

Peace,

James

Maybe it'll be better to have a little wider card (about 3") since I'll try a single layer. That way we can mount 4 channels on 1U and 8 on 2U chassis... I feel there's no need to squeeze that much... :D

I'm not so sure I agree with the concept of making things as tight and compact as possible. Especially with having so many pieces that you'll have to mount individually.

No offense to Kris (he's done some really nice stuff) but having to mount a common gain block (2-4 screws) an input tranny (2-4 screws), output tranny (2-4 screws), input/ouput XLRs (4 screws) and then the gain control/switches and indicators on the front. That's a shiteload of drilling and screwing. Then you're gonna multiply that by N channels. Not to mention all those little runs of wire you're gonna have inside the thing. It just seems ugly to me.

Using a 2U case has some serious benefits especially for the beginer. You've got some extra vertical space and lots of extra front/rear panel space. Plus you've got plenty of air around the components for cooling. (not that this project should get all that hot)

I was thinking that the main board would have the gain block, support components and the output transformer on it. That whole assembly gets mounted with 4 screws. The I/O module has the relays, XLRs and input transformer. It ideally would mount via the 4 screws of the XLRs. A small L-bracket might be necessary to add support for the weight of the transformer if you're gonna be lugging this unit around. The control board has LEDs, switches and a pot (or a gain switch) and is mounted via the hardware for the switches.

That's significantly less drilling of the case. I'd say that you can still run pretty compact with it too. I don't think you'll quite get 16 channels in 2 spaces but then do you really NEED 16 channels of this guy? Of course I guess you could always make different variations in the same box.

Peace,

James

Hi James,

The 317/337 regulators are a good way to go, but if the output cap is increased (the 1uF in the diagram above) then a protection diode will have to be placed from the "In" terminal to the "Out" terminal. For example, in the +Ve side, a 1N400x can be placed so that its' cathode is at the "In" terminal, and the anode at the "Out" terminal. This allows the cap to discharge through the diode rather than the '317 regulator! Don't forget to change the diode polarity for the 337!

Check out the datasheet for more info, p6 and 7:

[="http://www.national.com/ds.cgi/LLM117.pdf"]Nat Sem 317 Datasheet PDF[/]="http://www.national…"]Nat Sem 317 Datasheet PDF[/]

and:

[[url=http://="http://www.national…"]Nat Sem 337 Datasheet PDF[/]="http://www.national…"]Nat Sem 337 Datasheet PDF[/]

and some 317 Apps:

[="http://www.national.com/an/AN/AN-178.pdf"]Nat Sem LM317 Apps Guide PDF[/]="http://www.national…"]Nat Sem LM317 Apps Guide PDF[/]

Also, a protection diode from the "Adj" terminal to the "Out" terminal prevents the "Adj" bypass cap from discharging thru the regulator.

These aren't essential, but I think that the bypass cap on the "Adj" pin is essential for low noise, and if the output cap is going to be as much as 1000uF, then a In/Out diode is vital to happy regulator functioning.....I've accidentally discharged a big cap thru a 317, and I found bits of regulator sprinkled throughout the chassis :D

And watch out for the phantom- if taken from the DC side of the +/-25V supply it won't be suitable for +48V, because the +48V takes its' ground reference from the preamps 0V rail, which would be half-way between the +50V available from the +/-25V rails.

A good method was posted over in the Phantom Power thread today:

[[url=http://="http://www.forssell…"]Forssell Mic Pre PSU PDF[/]="http://www.forssell…"]Forssell Mic Pre PSU PDF[/]

...which uses a voltage doubler from one side of the transformer tap to stop ground-reference problems.

Hope this helps!

Mark

--------------------------
"Oscillators don't, amplifiers do....."
Anon.

No offence taken James...

My personal preference would be to have the input transformer located right near the gain card rather than at the XLR's, because the circuit unbalances the mic signal after the input transformer, and I'd prefer to have it balanced as long as possible (I wire my chasis with twisted shielded pairs).

You're right about the amount of drilling though. To have each component mounted seperately means lots of holes, and lots of hardware (more weight), but it's my prefered method of construction, as I never trust heavy objects (like transformers) to sit on circuit boards.

Cheers,

Kris

I hate to put water on this fire, BUT.
You folks are never going to agree on a mechnical layout
to suit everyone here. If you do the project
share the drawings, board layouts and such with the group.
There will be people to decide to jump in on your design / chassis / pcb layout
and others that design / chassis / pcb layout is not their cup of tea.
Too bad if we all got together on the board layout in the quanity
we are going to order the board would be dirt cheap. But that will not happen as
everyone wants somthing different.

adrian

That's right, it'll be very hard to suit everyboby, but it's always good to have some inputs from other folks. Actually I came up with a placement that seems to me very interesting, based on all your inputs. The only thing I won't include is the edge connector, that is not great to DIY.
If the board end up double layer and we send it to fab, I can include the edge on the same design. Other that this I think nothing is better that a good pad for a wire. I'll try to set up a homepage tonight to up the API stuff so you can take a look and tell what you think.

Hey Fabio,

If you zap the stuff to me I can add it to the 312 page that I've started. Just a thought.

Peace,

James

That would be a lot better... :D lemme scan the my draft or do a first placement and I'll sent to ya.

http://recording.or…

Hello,

I've deciced to try some html stuff, can anyone see this:

[[url=http://[/URL]="http://www.thediypi…"]API312[/]="http://www.thediypi…"]API312[/]

If you cannot see send me your email so you can host it...

[ November 12, 2003, 07:03 PM: Message edited by: Fabio Bauman ]

Originally posted by Fabio Bauman:
Hello,

I've deciced to try some html stuff, can anyone see this:

http://www.bauman.kit.net/api312_p1.pdf

If you cannot see send me your email so you can host it...

This link doesn't seem to work for me.

Steve

Originally posted by James Greenlee:
I think I tackled the OpAmp thingy awhile back. I s'pose that's in another thread. Anyway...The choices that I know about are:

Hardy JH-990
Millennia Media MM99
Forssell Technologies JFET-990
Avedis' 1122
Joe Malone's JLM-990
API 2520
Jensen 990

(Inward Connections SPA690 <= added from Joel Cameron's comments)

Anyway....For the purposes of creating an API 312 clone I would think that first up would be genuine 2520 opamps and second would probably be the 1122 opamps. I wouldn't know exactly though because I've never used them.

So there's some stuff to marinate on. Have fun!

Peace,

James

Has a decision been reached regarding which opamp to use? Other than the obvious choice of a genuine 2520, have we determined which choice gives us close to what we want for a reasonable cost? I made some calls today and so far it looks like the API2520 sells for around $90 each and the Millennia MM99 sells for about $75. Is there any way to do a group purchase of these to help keep the cost down like we did for the output transformers? I know there have been discussions and differing opinions on the input transformers and these opamps. I know the initial thought was to let everyone make their own choice on those two items; however this makes it difficult for us to do a group purchase. In the interest of saving money, it would seem to me that we should at least try and decided on a standard for the Group DIY project that is a price performance based selection. This way those of us who choose to go that route might be able to purchase them in bulk and everyone else who wants to customize their own design can gracefully abstain.

Regards,
Steve Ginn

ARGH!!! Too many API threads!!! (He he he...That sounds like the scream from a frustrated programmer more than a DIY'er.)

Check the other thread Steve. The one that's supposed to be in bold. I'll summarize here though for the sake of continuity.

Three main variants:

Current/Traditional : 1122/2520 opamp with Jensen or Reichenback input transformers.

Modded: 1122/2520 opamp with Lundahl 1538XL transformers.

Clean: JFET-992 opamp with Lundahl 1636 transformers.

My choice is 6 of the modded and 2 of the clean.

The price of Jensen input transformers drops about $20 a piece when bulk ordered in quantities of 48 or more. I can't seem to find much info on the Reichenbach transformers and haven't seen any price discount information on the Lundahl transformers. K and K audio seems to be the sole US Distributor for Lundahl. I don't know if you could order directly from Lundahl or not.

In related news...The order for the output transformers goes in tomorrow. I'll be ordering a few extras but not many.

Peace,

James

Sorry James for the cross post ... I actually posted the question here about the opamps and bulk purchasing before I read your reply in the other thread. Personally I like this thread better because it seems to have the bulk of the discussion and everything is in one place. I guess the other thread was started to try and distill things a bit, but this thread should be where we summarize all the findings and decisions.

$20 each for the input transformers sounds pretty good to me. Much better than the $50 price tag that was initially discussed. Do you suppose we might be able to get the same kind of deal on the opamps as well?

In your opinion, what are the characteristic differences between what you refer to as the "mod" version and the "clean" version? Which one do you think will be closer to the original? I am certainly interested in variety, and if the group purchase quantity requirements are less than they were for the output TX (100+) then we should be able to do a couple of combos and hopefully please most the those who ordered the 175 output TX.

Great work James!

Thanks,
Steve Ginn

$20 each for the input transformers sounds pretty good to me. Much better than the $50 price tag that was initially discussed.

Whoa!!! Slow down boy!!! Not $20 each...$20 off! The Jensen 115k transformers are about $75 a pop in singles. When you hit 48 the price drops to around $55. Just wanna clarify that!

Do you suppose we might be able to get the same kind of deal on the opamps as well?

This is highly doubtful. The discount probably just is NOT gonna happen. I think you need to keep in mind that the guys that make the opamps (like the 1122 and the 2520) use these for EXACTLY what we're trying to do here. Build PREAMPS! So essentially we're asking them to not sell us preamps and instead sell us the components.

In your opinion, what are the characteristic differences between what you refer to as the "mod" version and the "clean" version? Which one do you think will be closer to the original?

Well...The "mod" version is gonna sound very much like the OSA model with the Lundahl. To quote them:

"The spine rattling bottom end is huge, round, and full. the midrange is so present that you will swear that it can’t be solid state, and the top is elegant and silky."

Sounds good to me.

The "clean" version isn't going to have much character. It's more "wire with gain" that happens to have a little iron in it. The amorphous strip core Lundahl transformers are supposed to be smooth and liquid like nothing you've ever heard.

My intended use of this project is primarily drums. Although I'd like to use a Neve flavor on kick, for now I'll use the API clones. The clean channels are going to be used for OH's. Now obviously these pre's are good for much more than that. I would imagine that acoustic guitars through the clean channels would be stellar. Vocals would probably be nice through either depending on what you're going for. It just depends.

I s'pose a couple of extra channels of the more traditional design would be nice but after my initial 8 I'll want something different. (I'm kinda eyeing those Telefunken pre's that seem to be popping up everywhere.)

Peace,

James

What do you guys think would change the characteristics of the sound more? Input Transformer or Op-amp? Or both equally? :D Being the circuit is so simple we'd most likely hear every little component change. I'd like have a few different 'flavors' so to say.. I plan on building 8 channels of them. Six of them will most likely be the lundahl 1538 w/ the Avedis 1122. Not sure what to do with the other two.. Decisions.. Decisions....

-Jay

Who is the distributor for the Avedis 1122? Is there a web site?

Thanks,
Steve Ginn

Does anyone have the definitive 2520 opamp cct? I would like to get as close to the original but with our dodgy exchange rate it is very expensive to buy a genuine unit. The one cct we have seen had various notes to the effect of "extra stuff added in later versions". Any advance on this?

Peter

Its been a while since this thread has grown, so here goes..

I was wondering if those of us who have built this project could list what they did layout wise and what components they choose.