API Mic pre (with Yamaha parts) ***COMPLETED***

[mattmoogus]

[quote author="khstudio"]Matt,
Thanks man :thumb:

Do you think 50VA is too much?

Also I'm using the NE80200 if that makes a difference.

I'm also building that PM2000 2ch rack & need one for that... I was thinking about 1amp for both strips + Phantom = 24VA.
[/quote]


Sorry, I meant 80200.

1A for the 2 PM2K channels should be heaps. My PM PSU is 2.5A per 24v rail, and thats driving 70+ opamps as well as allll the IC amps aswell.

Just at a wild guess Id say you wouldnt need bigger than a 30VA transformer but listen to someone who knows about this.


M

 

[Greg]

Measure current on the V+ and/or V- rails, post regulation... ie what you're feeding the modules.

 

[khstudio]

[quote author="Greg"]Measure current on the V+ and/or V- rails, post regulation... ie what you're feeding the modules.[/quote]

:thumb: Thanks,
I was going to try that but wasn't sure if I had to double or half the reading I got because of the 2 rails + the link posted earlier to AVEL's site/info only mentioned AC... that's why I was confussed.

I really can't thank you guys enough for dealing with all my questions :grin:

As I said before my good friend, that was a great tech, died not to long ago & the only guy that's left around here that's any good doesn't know as much as he thinks he does. I can never tell when he right or wrong which throws me for a loop :roll:

Kevin

 

[khstudio]

Need some help... that's a surprise I'm sure :green:

Anyway, I'm going with a stepped gain (12 position Lorlin make before break)

I'm trying to determin the gain steps but since this is not exactly like the API, I'm not sure I can use the resistor values posted by others.

My input tranny is 300:2.7k - so theres not as much gain up front as the API, which doesn't seem to be a problem because I have what apears to be about 62dB total (if my testing was correct)

Output tranny is 600:600 - no gain here.


:idea: What I've done so far is to compare the gain (dial) settings to a pre I'm used to (MP20)... like where I set it for voc, guitar, u87, 57,58, ribbons...

The 2nd part of my tesing:
I have a cable tester that will put out -50 mic level.
Running thru my MP20 & viewing the output in WAVELAB:

Gain full CCW/off = -44dB
Full CW/MAX = -2dB

API/YAM:

Gain full CCW/off = -22dB (louder)
Full CW/MAX = WAY OVER!!! Couldn't measure

Obviously has more gain than the MP20.

API/YAM VALUES:
(R3) FEEDBACK = 22K
(R2) = 100
GAIN POT = 25K

To sum up:
I was just going to use the -50 test tone, set the gain on the MP20 to known setting, view it in the computer, apply settings to the API & measure the pot to make sure THOSE values are incorporated into my stepped gain.

BUT, Even with the 25k pot turned down it seems to high.

Can anyone help?

Kevin

 

[Greg]

I just thought about this for a second, but how about this approach:

1. Put a pot in there for gain control.
2. Determine the max gain... maybe 50-60dB? I don't know really what the Yamaha can do and stay stable...
3. Determine the min gain you want to run the preamp...
4. Divide this range in twelve positions...
5. Put in a known input signal, let's call it Vin...
6. Measure the output to get the gain (in dB) that you want, using this equation:
Gain (in dB) = 20* log10(Vout/Vin)

For example, let's assume we're trying to find the 35dB gain resistor and our input signal is 20mV=0.02V.
35dB = 20*log10 (Vout/0.02)

Solve for Vout:
Vout = 1.13V

Tweak your pot to measure 1.13V on the output (with 20mV input). Disconnect pot, measure resistance, and presto. Do this for all 12 desired gain points.

I'm sure there's an easier way to do with only calculation, but this "trial and error" method should work fine.

 

[khstudio]

Greg, Thanks for staying with me on this :thumb:

I think I got a grip on what your saying & how to do the calculations.

For some reason, I can't get my output gain to go down that much compared to the MP20. (See last post)

I also found that the input tranny, when wired like Yamaha had it, was flipping the phase... so I flipped the input instead of the output & it's sounds a lot better for some reason. Remember earlier I posted about the voltages I was seeing @ the output (pin 2 & 3) being different... well after testing I found that the one that measured higher was the + or positive swing so when I originally flipped the phase there (to match my other pre's phase) it didn't sound as good.
Could this be from the OT being or causing it to be asymetric? Just guessing here...

I also listened with & without an output CAP & DEFINITLY like it better without it. Wasn't bad with it - slightly constricted sounding but lifted up a little bit or honk & rolled out highs & lows. (I used a Pan FC 220uf for testing)

I hope my Trannys can handle the little bit of DC I'm getting - Tranny resistance is 20 ohm & the DC is about 20mv. CJ or someone mentioned some math involved to get the actual DC on the OT & said it would be much lower than the raw DC @ the Tranny input. I don't really know what I'm doing with regards to the DC though.


I still have a little bit to go but it's does sound GREAT.

Thanks,
Kevin

 

[Greg]

[quote author="khstudio"]For some reason, I can't get my output gain to go down that much compared to the MP20. (See last post)[/quote]
Don't even worry about this. You're the designer here. Determine the min gain YOU want to run the preamp, and set it there. I think rafareed posted some equations on how to do this some pages ago.

[quote author="khstudio"]I also found that the input tranny, when wired like Yamaha had it, was flipping the phase...[/quote]
The best way to determine if polarity is being flipped is to use a dual trace scope. Is this what you used, or were you using your ear?

I can't comment on your output tranny assymetical issue.

 

[khstudio]

[quote author="Greg"][quote author="khstudio"]For some reason, I can't get my output gain to go down that much compared to the MP20. (See last post)[/quote]
Don't even worry about this. You're the designer here. Determine the min gain YOU want to run the preamp, and set it there. I think rafareed posted some equations on how to do this some pages ago.

[quote author="khstudio"]I also found that the input tranny, when wired like Yamaha had it, was flipping the phase...[/quote]
The best way to determine if polarity is being flipped is to use a dual trace scope. Is this what you used, or were you using your ear?

I can't comment on your output tranny assymetical issue.[/quote]

Well, I think my test was flawed... I'm using this Beringer cable tester that will put out +4, -10 & -50 (mic level signal). I have no idea what's going on.
I'll try to explain:
I WAS taking a mic (57, u87, etc...), setting a good level on the MP20, then aplying the -50 level in place of the mic so I could then take the measurments & apply them to my new pre. The presonus reacted fine.
-BUT-
but when I apply the -50 signal to my new pre it was WAY TOO loud. I don't know if it's impedence or what. I'm going to try a different source.
This COULD be the cause of the gain appearing to be not low enough (like I mentioned in my last post) - cause when I put a mic on it it's WAY TOO low... so somethings wrong with the test. I can say that the signal comming out of the Beringer tester looked like HELL on the scope - junk, junk, junk.
----------------------------------
Testing Phase:
I've been splitting the test tone signal out of this Beringer cable tester -
One to my console & one thru the new pre & trannys.
When I find the one that cancels out, as oppossed to adding gain, I know I'm out of phase.
---------------------------------
A lot of the calculations are new to me & like others, I haven't had Algabra for over 15 years, I'm 36. So I've been doing nothing but reading for a few days to actually learn what I'm doing. Maybe I'm blind but I can't find how to calculate my input tranny gain or ratio.
It's 300:2.7k

Thanks,
Kevin

 

[khstudio]

Greg,
I know you were thinking about the PM1000. I'll tell ya, it's a KILLER pre & an easy & cheap project. I love this thing on vocals - 58, u87. There's something to the 0 to 44v (Single rail) that has a different/cool feel than other pre's I've heard. Someone mentioned in another thread how the single rail clips (rolls over or whatever) a lot smoother than a Dual +/- rail setup.

I hope I worded that right. I would be happy to help & share ALL my notes with you.
Dave/Soundguy has a few for sale :wink:

Kevin

BTW, I love passive EQ's & it's got that too. Very musical & adds a nice uplifting, un-annoying color to the track. A cool added bonus to a great pre.

 

[khstudio]

OK, the beringer tester is wack

I'm using my computer now - the signal is very balanced on pins 2 & 3, unlike the beringer = nothing from pin 3 and ALL signal @ pin 2.
I think this is PART of the bullshit I was trying to explain about 3 or 4 pages back.
--------------------
From the measurments I've made concerning the input tranny, it looks like about 6-10 dB of gain (300:2.7)
Does this sound right?

I'm trying to determin the gain the Input trannys adding so I can remove it from the equation for the stepped gain.
+ I've never done this stuff before... so I'm just poking around & learning what's going on.

Kevin

 

[rodabod]

[quote author="khstudio"]Maybe I'm blind but I can't find how to calculate my input tranny gain or ratio.
It's 300:2.7k
[/quote]

Kevin,

That figure is your transformer's impedance ratio - 300 Ohms at the primary will become 2.7K Ohms at the secondary.

Your impedance ratio is 2700 / 300 = 9

Your turns ratio (voltage ratio) = Sq.Rt. of Imp. Ratio = Sq.Rt. 9 = 3

Therefore transformer turns ratio is 1:3.

Transformer Gain (in dB) = 20 * Log( Sq.Rt.(sec. impedance / pri. impedance))

= 20 * Log 3 = 9.5dB gain.

I better read that again later, as that isn't the sort of ratio I expected for an API pre. Perhaps someone else can chip-in.

 

[khstudio]

Thank you for helping

I know I saw that formula around here but it's nice to see how it works with my own stuff/tranny... helps me understand better.

YES, you are correct - this pre-amp is based on an API 312 but with my Yamaha discrete Op-amps & Trannys (& their Zobel).

Input Tranz 300:2.7
Output Trz 600:600

Other Changes:
R3 = 20,000
R2 = 100

Which gives me
+46 dB ( Op-amp Gain )
+ 9.5 dB ( Input Tranny )
= 55.5 dB ( TOTAL gain )

I know I'm gonna get reeeeemed for asking but :
What is & where do you get the "20" & or "LOG" :?:

Is LOG a certain # like 1.414
I DO understand how you got the turns ratio of 1:3 - but after that I'm lost.

I'm currently trying to calculate my stepped gain -

I've been using this thread to help me learn:
http://www.groupdiy.com/index.php?topic=40518&highlight=api+stepped+gain#40518

Along with this dB calculator fro AD:
http://www.analog.com/Analog_Root/s...ols/interactiveTools/dbconvert/dbconvert.html

& this one when using my volt meter:
http://www.sengpielaudio.com/calculator-db-volt.htm

20,000/100 = 200 (46.02dB op-amp gain)

Once I learn this crap I'll probable slap myself for how easy it is... but til then :guinness: :green:

 

[rodabod]

[quote author="khstudio"]
I know I'm gonna get reeeeemed for asking but :
What is & where do you get the "20" & or "LOG" :?:

Is LOG a certain # like 1.414
I DO understand how you got the turns ratio of 1:3 - but after that I'm lost.
[/quote]

The 20 and the Log are used for converting a normal ratio into Decibels.

You know that dB are a good way of expressing a ratio or gain, so we use these here, otherwise things would get out of hand (eg. talking about a preamp set with a gain of 100,000.....)

Log (Logarithm) is a function / law. You know what a logarithmic potentiometer is?

Go and search for logarithm on the net and try and find a graph representing one too.

Now, 20*log is part of the mathematics required to calculate dB, and this can be easily proven, but I think if I were in your case just now, I'd just learn the formulae and use them.

This may help:

http://hyperphysics.phy-astr.gsu.edu/HBASE/sound/db.html

But remember when to use either 10 or 20 as a multiplier as it differs for power or voltage. Note how when you describe a value in dB it stays nice and small!

 

[oobedoob]

Does anyone have any more info on the differences between 80100 and 80200 opamps? (datasheets?)

I like the idea of this project but I only have 80100s (about 8 of them and I want to put them to good use)

 

[mattmoogus]

Actually I pulled some of them out of my desk last week and we took pics of them. The 200 is potted with a clear gel type substance and has 8 transistors and the 100 is potted with black epoxy and you can see the metal tabs of two large transistors poking through. The 100 is the output driver opamp and clearly from its construction it has more current drive than the 200.

What info do you need? You could always just put one into a circuit and see what its like.

The 200 may have lower noise than the 100 but were not sure. The best way is to put it in circuit and try.


M@

 

[oobedoob]

[quote author="mattmoogus"]
What info do you need? You could always just put one into a circuit and see what its like.

The 200 may have lower noise than the 100 but were not sure. The best way is to put it in circuit and try.


M@[/quote]

That's the plan for this afternoon. I suppose I could unpot one as well...

is there a meta about that?

 

[khstudio]

[quote author="oobedoob"]Does anyone have any more info on the differences between 80100 and 80200 opamps? (datasheets?)

I like the idea of this project but I only have 80100s (about 8 of them and I want to put them to good use)[/quote]

If there are data sheets around I'd like to have them too :wink:

There's a guy on the net, Fletcher (famous for some reason, I don't know what but he seems to know what's up) I had read that he put the NE80100 in a real API pre & thought it sounded great.

I started this thread not fully understanding what I was doing but had all these great parts. Now I realize that the API, circuit wise, isn't that different from other mic pre designs & that these discrete opamps arent worth a fortune but do sound very good. JLM & others sell op-amps for about $50-60 each, so having 8 saves you a little cash there. You also don't have to use an OT. I'm squeezing a little more gain out of mine because my input tranny is 1:3 & I think the original API is 1:8 or 1:10. But it sound great to me. JLM hase the DUAL 99v boards & kits available - I would have bought them but they weren't sure about the NE80200 (that I have) but they WERE sure about the 80100, like you have. & say it will work fine. One downside is they DO NOT sell a kit WITHOUT the Op-amps. I'm not sure why... they do offer just the boards but you''' have to order all the other parts yourself. I would have DEFINITLY bought the kits, less op-amps if they offered them & if I was certain the "200's" would work... but your there with the 100's.

BTW, I designed a circuit board for my API/YAM that I will post soon. I couldn't really buy one pre-made because I'm using MOST parts from this Yamaha M1516 console I've got & the Yamaha op-amps have SQUARE PINs... most, if not ALL other op-amps have round pins.
Also, tha Yamaha switches have a larger footprint & won't fit the other boards available + I saved a few bucks there as well.

I MAY end up with some extra parts to sell but I'm not sure yet...

Let me know if you need any more info & I'll try to help.

Do you have a console or just the amps?

Kevin

 

[khstudio]

I've noticed that a lot of the NEWER API 312 designs use larger caps for the +/- Rails (op-amp power Filtering) like 220uf instead of the 12uf original.

Should I increase this as well :?:
I noticed my PM2000 uses 220uf here along with a .1 mono cap.

Also, I noticed newer designs use a higher value for C3 (250uf original)
I originally bought 220uf for this cap but I do have 470uf available... Could someone explain what a higher value would do @ this position & if I should increase mine also :?:


Thanks,
Kevin

 

[rafafredd]

I would use at least 1000uF. It´s for flat frequency response on all gain settings. You could go as high as 4700uF.

Also, 12uF is way too low for the opamps bypass. Electrolytics were much more expensive by the time that they drew that API schem.

 

[khstudio]

[quote author="mattmoogus"]I really should ask Joe about this, I may be wrong but...

Id say all you need to load the xformer like they do is the 150k and 2K7 + 470pF zobel across the secondary.

Id also try it without the zobel, loading it with 2k7 and 10k and seeing what it looks/sounds like. If you find it sounds good with other loads, you might include a load switch for two different tones.

In the PM2000 it sounds pretty well damped, the circuit from mic pre in to insert out is about 0.7db down at 20k with a nice smooth gentle roll off.


M[/quote]

Matt, I'm stuffing the PCB's now, loading each one a little different so I can hear the differences & test them with the RMAA app.

Couple questions:
Zobel, You mentioned "loading it with 2k7 and 10k"
Could you be more specific as to where.
I wish I tested this before making my boards... the switchable Zobel (like you metioned for different sounds) seems like a great idea.
------------------------------
:EDIT: I found your detailed responce earlier in this thread about the Zobel/Loading & 680 ohm output loading (for testing)
------------------------------
I feel there's just a little too much high-end roll off after tracking a kick & snare yesterday...I'm NOT using a cap on the output & I must say I've NEVER heard subs on my kick or a phat bottom/punch/warmth on my snare, ever like this. I guess it's from not having caps/electro's in the signal path???
Now, I'm going to test/listen again with a cap to see BUT...
Concerning the high end being a little too rolled off, I'm wondering if the small amount of DC could be causing it or if the increase in SUB's (that I'm not used to) is affecting my perception of the highs... I don't know?

Is it possible that you (or anyone) could explain what's doing what on my Zobel so I could inderstand what's happening when I increase or decrease values or even remove them?

My current ZOBEL:
150k - across secondary
then - 2.7k & 470pf across secondary

---------------------------
RMAA Testing:
Just to make sure I'm in line here, I built a U PAD to run from the +4 out of my sound card to my pre's=
2k-2k/300

I'll also increase the bandwidth for the next test...
Along with loading the output (like you said) but just to make sure... exaclty what & where on the output?

Thanks,
Kevin