API 312 Thread!

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
MountCyanide said:
This OLD thread has been so helpful.
With this new addiction to DIY and my strengthening tolerance I had to up the dosage.
Ordering PCB's wasn't doing it for me. So I broke out the Ferric Chloride and learned to etch.
Thanks to the archive of this board and all of the brilliant and curious minds I pulled it off.
4 Channels of 312 preamps. And they sound awesome. Cinemag CMMI-8 in 2 and 75101 in the others.
Pictured above them in the rack is the Whistle Rock ML-12 which started this hunger for more API's.
What a satisfying feeling to make these things by hand. The etch really adds another layer of euphoria.
Not only the fumes of poison either.

Beautiful build! Check your mailbox? I just pm'ed you!
 
Hi,

just starting to put together 4 x API-312 style pre amps using these pcbs, i'm going with THAT1646 output and Sowter 4257 on input TXF.
Have sorted the psu and have good voltages +/- 18v and 48v.
This is the BOM i've put together just wanted some verification and everything really so i'm not putting bits in that i don't need.
Only going to have 48v switch, Pad switch, Polarity switch.

Resistors

R1/R2/R3 = 3.9K
R4 = 200R
R5/R6 = 6.8K
R7 = 170R
R8/R9 = 620R
R10 = Zobel Input TXF
R11 = 1K *
R12 = Input TXF load resistor (100k-150k)
R17 =
R18 = 200R
R19 = 47R *
R21 = 47R protects OPAMP
R24/R25/R26 = 383R
R27 = 3.9K ouput pad led *

Caps

C3/C4/C5/C7 = 0.1uf Wima
C6 = 330uf/16v Bipolar
C8 = 470uf/16v Bipolar
C9/C10 = 10uf/50v Bipolar
C18/C19 = 1000pf *
CF = 100pf
CP = 68uf/63v
CZ = zobel cap

Diodes

D1/D2 = 1N4004
D3/D4 = 1N4148

IC2 = THAT1646 or DRV134

Pot = 50K Log?

not sure on the pot but can i use the same as i would with the other 312's, 20k or 10k i think?

regards

Spence.
 
Hi,

Was hoping for some advice on what to specifically leave off the pcb.
Do I need to put in the 1n4148? Or the 1n4004?
I'm only using Sowter 4257 input txf and line driver 1646/DRV134.
Should I just refer to API-312 schematic?

Also I have some stepped pots which I have made into 25k rev log , but these switch pots are not continuous connection, should I expect thumps when switching through the gain, or is there a way of fixing this? Alternatively I can just buy the 22k rev log pots.

Also do I need the input bias section?, gain adj? DC servo?
Was thinking I could just bypass these sections

Regards

Spence,
 
Spence,
Please post in the thread I linked to in my last post, otherwise, people who have built the PCBs you are using may not see it and be able to help you.  There have been a million different 312 PCBs and this is the old general thread. Good to read through in any case, but if you want info specific to a project go to that project's thread. 

Here is is again:
http://groupdiy.com/index.php?topic=41339.0

The BOMs, schematic and all other info is there.  As for what to stuff and what not to, all sections are boxed off in both the schematic and PCB silkscreen.  You only stuff whatever you need/want.  For info on bias and how to, go to John Hardy's site and download his 990 and M1 papers, there is a lot of great detailed info in both of those. You will get more than you need for sure.

 
Hi,
this is a really big thread I'm afraid my question will get lost, but anyway this is the proper place to ask so I hope you can provide your opinions.

I have some 312 Mic Pres to build and would like to know your opinions on Cinemag (CM-75101APC, CMOQ-2S)  VS  Ed anderson (EA2622, EA2503) for this circuit.

There's any preference in sound between Cinemag or ED Anderson transformer?
What do people feel are the sound differences or characteristics of one or the other?

thanks


 
I would like to add a Line Input to an API312 Clone. I've seen how Fabio made it. He used a relay for switching, but it should work with a switched jack as well. Or would it be a problem because of phantom power?

Another question:
The most people here are using Bo's DI Mod. I cannot find much about an post 2622 DI. Has anybody an idea?

 
So a while back HandcraftedTone posted some Eagle files for the 312 card. I want to make more, but can't seem to find the files and have only found dead links. Do any of you have the files?
 
I'm wondering if someone can lend me advice on a couple of input tranformers. They are from an Altec 1674 mic mixer. The only instance I can find reference to them is here: https://groupdiy.com/index.php?topic=44710.20 where Bruce O says:

"Altec 1674 1678 1684 automatic mic mixers have some marked ALTEC/671-1323/651-8957 in them, square Mu Metal can, center tapped, 1:4 and 1:8 ratios. Tested pretty good, well shielded. These mixers also have a line output transformer in them (in the case of the 167x it is the same size as the api output trafo's I have not tried it though)."

I've attached the best pic I can find of the input to 1674(8) mixer.

There are 3 primary pins and 3 secondary pins. I believe in relation to the mixer schematic, that means the middle pin on both sides is the center tap, forcing me to series wiring.

I'm getting 12 ohms between the center tap primary and both the other windings on the primary. So, 24 ohms between the outside windings. The secondary has the bottom pin grounded to the case. Between the top pin of the secondary and the center tap I'm getting 1.3K

So question is, can I use this for the API circuit? And if so, what do I do with the center taps on either side?

Building this for a buddy for Xmas.
 

Attachments

  • Screen Shot 2020-12-09 at 1.12.30 PM.png
    Screen Shot 2020-12-09 at 1.12.30 PM.png
    299.9 KB · Views: 30
JW said:
There are 3 primary pins and 3 secondary pins. I believe in relation to the mixer schematic, that means the middle pin on both sides is the center tap, forcing me to series wiring.

I'm getting 12 ohms between the center tap primary and both the other windings on the primary. So, 24 ohms between the outside windings. The secondary has the bottom pin grounded to the case. Between the top pin of the secondary and the center tap I'm getting 1.3K

So question is, can I use this for the API circuit? And if so, what do I do with the center taps on either side?

API typical 2622 transformer has an 1:7 turns ratio.
You have to see if you can get similar turns ratio with those transformers, if you got 1:8 that would be pretty good.

In the schematic you posted it shows 2 separate primaries and 2 separate secondaries, that have Pin 2 and 1 and then 5 and 4 connected in the middle to make a center tap.
Can you remove the center connection between pin 1 and pin 2 in the primary?

I would imagine something like Primaries in parallel and secondaries in series. But better to calculate the turns ratio to be sure and I don't remember how it's done.

Post a separate thread on this subject in the "Magnetics" section of this website you will get more specific info on transformers over there, as this thread is related to the API 312 and that transformer was never used in the 312 so people here might not know hoe to help.

best regards
 
Okay, I'd like to link this thread if anybody is interested: https://groupdiy.com/index.php?topic=15010.0 which involves building 312's out of Yamaha parts.

My question is how low can one go on R2? Does it induce lots of distortion if it's too low or something? The reason I ask is I built up 1 channel with the usual EA/CAPI transformers,  200 ohms in this spot (R2) per usual. I was easily able to achieve the same gain using Yamaha parts on the other channel. I was using a PM1000 input transformer wired at 150:2.7kohm stepup. It's terminated with a 4.7K resistor in the PM1000 strips, so that's what I used. No zobel (yet) I used Yamaha 600:600 outputs as well. And a Yamaha 80100 opamp.  I just put a trimmer on R2 the resistor on the way to the volume pot before the volume pot's cap. Is this okay and all that's required? And I can use a lower ratio input transformer?

Also, C4 the feedback cap I see Kevin (from that thread above) changed it to 47pF like Yamaha used with their opamp. My question is, is it better to adhere to the Yamaha's schematic or API's if I'm using Yammy parts?
 
Answering some of my own questions with a link where Harpo and Bo Hansen explain the feedback resistor and cap. https://groupdiy.com/index.php?topic=26948.0

I just measured 37ohm as the value I needed to obtain 'normal' 312 gain using these Yamaha parts. That is for the resistor in series with the volume pot.

So my values are:

120pF feedback cap
20K ohm feedback resistor
37ohm for the above mentioned resistor in series with the electrolytic for vol. pot
470uF for the lytic.
22K rev audio pot.

In the thread linked above Bo Hansen says: "Of course there are a disadvantages to go too low or too high with the gain potentiometer value, and this because the 2520´s behavior and also a components choice concern size vs quality.
Too high, 100 kohm and up, can result in to much noice, because 2520 have bipolar transistor inputs.
Too low, 4,7 kohm and down, need big coupling capacitor to ground, and can maybe result in scraping/thumbing gain pot.
So for the 2520, I think the 22 kohm are a good average value."

Anyhow, also mentioned are several relationships to achieve the same LPF and HPF's and gain for the 312 but with different gain pot.

For example, a 4.7K pot (the lowest you can go?) is used with a 4,7k/470pF feedback circuit, and 47ohm for the series resistor and 1000uF for the series cap.

This is more in line with what I'm measuring (37ohm series resistor) for the gain I like, but then the other components 'conspire' to bring the gain down correct? (corresponding to the 2622 input ratio and not the Yamaha transformer I'm using)

So, can I simply leave the 120pF and 20K feedback circuit and keep this 37ohm resistor on the way to the cap and (22K) volume pot without effecting freq. response and phase rotation too much?

I see in the thread I linked last post that Kevin (KHStudio) went for a 22K volume pot, 47pF (like Yamaha)  and 20K (not like Yamaha) in the feedback circuit, and fudged the series resistor a bit, to (100 ohms)
 
JW said:
...I just measured 37ohm as the value I needed to obtain 'normal' 312 gain using these Yamaha parts. That is for the resistor in series with the volume pot.
This shunt arm resistor in series with the pot with rev.log taper, wired as a rheostat, set the amount of gain (NOT a volume pot) in correspodence with the feedback resistor. Max.gain is set with the pot dialed CW for zero ohms, so only the value of the fixed resistor (0 ohm + 37 ohm = 37 ohm) remains as the shunt arm resistor value.

So my values are:

120pF feedback cap
20K ohm feedback resistor
with pot value 22K, this feedback resistor value would be same value 22K.
Your 120pF cap now would set the LPF to 60,286Hz instead of 66,315Hz.
I'd probably decrease this cap value to 47pF for a 154kHz LPF and fit a maybe 27pF in parallel, if the opamp/DOA should oscillate/misbehave.
37ohm for the above mentioned resistor in series with the electrolytic for vol. pot
setting the max.voltage gain to 1+(22000/(0+37))=595.6 , this is log(595.6)*20=+55.5dB.
For usual this stage is set for a voltage gain of 100=+40dB. Dunno, if your DOA can do this amount of gain without misbehaviour.
470uF for the lytic.
setting the worst case HFP=max.gain to 1/(2*pi()*37*470E-6)=9.15Hz
22K rev audio pot.
the gain setting pot (NOT a volume pot).
This is more in line with what I'm measuring (37ohm series resistor) for the gain I like, but then the other components 'conspire' to bring the gain down correct?
the other components are the 22K rev.log pot.

(corresponding to the 2622 input ratio and not the Yamaha transformer I'm using)
voltage gain of the input and output transformer is on top of the gain stage. With 1:7 in and 1:2 out ratio, this is +22.9dB on top of your 55.5dB gain stage.
 
Awesome. Thanks so much Harpo. I think I'll sub in a 1000uF cap for the 470uf  to decrease the hpf to 4.3Hz, which is closer to API. Depending what kind of cap I use I suppose I could go even higher.

Do these start acting like antennas though if they're too physically large?
 
JW said:
I think I'll sub in a 1000uF cap for the 470uf  to decrease the hpf to 4.3Hz, which is closer to API. Depending what kind of cap I use I suppose I could go even higher.
context? 'Closer to API' would have a gain stage maxing out at +40dB. When you dial in your pot to same value as the API312 (37 ohm fixed + 183 ohm pot setting = 220 ohm), your HPF with 470uF in the opamps shunt arm is set for 1.54Hz. A source that requires this amount of gain (+40dB + 22.9dB transformer gain = +62.9dB) most likely isn't a line device but a mic. As long as you don't want to investigate seismic figures with this mic, you probably want this HPF some octaves up in order to get rid of FI stage rumble. As a makeup gain stage for a passive summer or passive EQ, most likely a lot  less gain is required wich in turn further decreases the HPF. IMHO I'd keep the 470uF. As always YMMV.
 
Thanks Harpo.

Application is a mic pre yes. I'd like the versatility to be able to record close miked drums or ribbon mics at let's say medium volume. I'm not asking to record distant ribbons on quiet vocals or something.

Anyway, you're deriving the 183 ohms from let's say, the pot mostly cranked but backed off a bit?

Concerning the hpf, If I'm running the numbers right, API's high pass  -3dB knee is at 3.18Hz w/ the 250uF cap and 200ohm shunt resistor. And the Capi schematic with the 330uF cap comes to 1.47Hz.

So, assuming full gain, I'm filtering more low end away than these designs even with a 1000uF cap. (4.3Hz)

Close to API I was just referring to the hi pass -3dB knee, not the gain. Though please school me even more on how they're related if you like. I'm just starting to learn this stuff.

Let me ask, if the distortion that occurs at higher gain stages with my preamp with the lower stepup transformers (55dB and up?) is pleasing, then are there any other ramifications of forcing the opamp to squeeze another 10dB out, or if I'm more conservative, I could back down to 60dB max.

I remember the Yamaha PM1000 (channel strip) is supposed to have a hard time pushing 600 ohms with it's two gain stages, but there is so much gain available when adjusting the negative feedback, and I thought it was remarkable how free of distortion (to the ear) this extra gain seemed in reality. But, that's another preamp.
 
Update. Okay, I tried out this Yomamma 312 on actual instruments/voice. There's quite the darkening in the upper frequencies (compared to my 1290) . And the gain range definitely is not usable past a certain point because of unpleasant distortion. So, guess I'm learning the hard way to use appropriate input transformers. (I switched the feedback cap to 56pF which (did?) raise the low pass filter. . .but I think the input Xformer should ideally be switched out)
 
Has anyone tried a switch to switch between series and parallel wiring on the input transformer primary as an alternative to a pad?

I see whey people talk sh*t about the pad. . . it does seem like it loses some life.
 
Back
Top