Yamaha m1516 NE5534 Mic Pre version attempting API 2520 Mod...

[stuartpre]

Hi All,

So i bought a Yamaha m1516, you know the one with Transformer on every mic input.

But I open it up and to my surprise, the op amps are 2520 footprints soldered in but they are NOT discrete. Instead they have a NE5543 op amp chip and some resistors, caps, transistors,  etc.

Further I have a API clone by CAPI at home that I put together a while back and dig the sound, it has 2520 op amp....

So I I compare the schematics between Yamaha and API 312 (CAPI) and they seem similar to me in design,

both non-inverting, gain control between the inverting input and earth, both with a transformer front-end.

Question 1. Now I ask myself, what happens if I drop in a GAR2520 from CAPI, will it work in the circuit and improve the sound or, should I be pulling out all Yamaha stuff and just duplicating the API 312 CCT?

Question 2: I notice the Yamaha has some series inductors in the audio path, can anyone tell me what they do? and do I need them?

Question 3:  I intend to put an output transformer somewhere, maybe on the insert send... post EQ..? But the API 312 has one following the 2520 op amp and is direct coupled with just a resister to limit current i guess. So what do I do to correctly impedance match and block DC if necessary between the 2520 output to the EQ stage?

To summarize; I wan't it to sound as good as I can, i'm going to mod the EQ with the JLM LCR style mod anyhow so in that context, what do you advise?

PS, i'm happy to get some CAPI output transformers if that will work better than the Yamaha's in the board available for re-purpose.

PSS Oh i've already ordered the CAPI summing and booster amp so it will have an API mix buss, and i'll retain  PRG Buss 34 as the Yamaha mix buss just for fun.

Schematics: copy link and paste in your browser

http://www.capi-gear.com/catalog/images/gallery/VP312/VP312-VPR-Rev-A-schematic.pdf

http://elektrotanya.com/yamaha_m1516.pdf/download.html

 

[Humner]

Sounds like it already has been modified.

It should have Yamaha NE80200 where those NE5543 are

Do you have any pictures? Kinda curious to see it.

 

[abbey road d enfer]

Hi All,

So i bought a Yamaha m1516, you know the one with Transformer on every mic input.

But I open it up and to my surprise, the op amps are 2520 footprints soldered in but they are NOT discrete. Instead they have a NE5543 op amp chip and some resistors etc.

That's kinda weird because the original 80200 is a pretty decent DOA that has nothing to envy to a 5534. In addition, the nominal supply voltage (+/-23V) exceeds the maximum ratings of the 5534 (+/-22), which would put a serious doubt to long-term reliability, unless some of the additional components you mention actually drop down the supply - which is also subject to question.

Question 1. Now I ask myself, what happens if I drop in a GAR2520 from CAPI, will it work in the circuit

Yes.

and improve the sound

I doubt very much you could objectively observe an improvement, but audibly, the possible differences may be to your liking... or not.

or, should I be pulling out all Yamaha stuff and just duplicating the API 312 CCT?

You could also do that, and that may also give an audibly different result, that you may or may not find to your liking.
Most of the character of the 312 comes from the interaction between the 2520 and the 2503 transformer. You would not have that, unless you add a 2503 xfmr in the channel.

Question 2: I notice the Yamaha has some series inductors in the audio path, can anyone tell me what they do? and do I need them?

I don't see them on the schemo, but inductors in the audio path are used for suppressing RF interference. If they are in, let them in.

Question 3:  I intend to put an output transformer somewhere, maybe on the insert send... post EQ..? But the API 312 has one following the 2520 op amp and is direct coupled with just a resister to limit current i guess.

The 1R resistor there doesn't do anything significant. It's a remnant of days when the designer experimented with optimum coupling between the DOA and the xfmr, and also offers te possibility to install a DC bocking cap for an unbalanced output.

So what do I do to correctly impedance match

Forget about "matching"; the concept here is "bridging", based on a very low source Z coupled to a much higher load Z. The ouput Z of the DOA is a fraction of ohm, and the minimum reflected Z at the primary is about 75r.

and block DC if necessary between the 2520 output to the EQ stage?

It depends of the total load that's presented to the capacitor. If the xfmr is connected directly to the primary of the xfmr and a signal is derived from the DOA to the EQ, the capacitor will see only the EQ input Z, which is variable, but never smaller than 2.5k. You could calculate the capacitor value using the usual formula C<<1/Z.2pi.f
with f being the lower frequency limit and << meaning generally 10 times smaller
resulting in C=1/(2500*6.28*20)= 30uF for 20Hz, which is consistent with the choice of a 47uF cap by Yamaha.
Now if you want to connect the xfmr after the coupling capacitor, the load value used in the calculation will depend on the xfmr (particularly its primary inductance and its turn ratio) and the expected load impedance at the secondary.

To summarize; I wan't it to sound as good as I can,

I would be tempted to answer: get rid of all unnecessary transformers, but it's just my biased opinion.

PS, i'm happy to get some CAPI output transformers if that will work better than the Yamaha's in the board available for re-purpose.

The output xfmr's in the Yam are 1:1; the CAPI's are 1:1+1 or 1:1+1+1. They can be connected as a 1:1 or a 1:2, even 1:3, but they have quite a low inductance, which may be difficult to drive for the line amps. I don't know the characteristics of the original xfmrs in the Yam, but I suspect they have quite a higher inductance. Experiment at your own risks...
Again, for xfmrs, nothing in the specs indicates what the sonic differences may be.

 

[stuartpre]

Thanks 'abbey road d enfer',

I see now that what I thought on the schematic was an odd inductance symbol, but be a symbol for a particular type of capacitor. Now that makes sense as you refer to a 47uF cap, that would be the 25 v one following the op amp, yes I see that now. So no inductors in the audio path, check.

You wrote "If the xfmr is connected directly to the primary of the xfmr", did you mean 'if the the DOA is directly connected to the primary of the xfmr'? Could you clarify that paragraph, I think i follow you and the calculations are making sense, i'm just sketchy on the possible configurations.

To be specific, what should I do if:

I want to drop in the GAR 2520 and follow with a 2503 xfmr (as per CAPI schematic) then go into the EQ  (leaving out the 47uF cap and 100r resistor) will that work ok? then i can take my tape send unbalanced from the insert send as usual.?

 

[abbey road d enfer]

You wrote "If the xfmr is connected directly to the primary of the xfmr", did you mean 'if the the DOA is directly connected to the primary of the xfmr'?

Yes, it's a typo.

Could you clarify that paragraph, I think i follow you and the calculations are making sense, i'm just sketchy on the possible configurations.

To be specific, what should I do if:

I want to drop in the GAR 2520 and follow with a 2503 xfmr (as per CAPI schematic) then go into the EQ  (leaving out the 47uF cap and 100r resistor) will that work ok? then i can take my tape send unbalanced from the insert send as usual.?

  There are several possible answers to that.
You could:
A) Just add the xfmr, connecting the primary right at the DOA's output; that would be the closest to the 312. Remeber that the offset voltage at the output should be kept to a minimum; that should be the case if you use a reasonable version of the 2520.
The signal going to the EQ should be essentially unaffected by the xfmr.
B) Add the xfmr right after the coupling cap, before the 100r resistor. You would need to increase the capacitor from 47uF to 470+uF. I don't recommend that. The only advantage is that it guarantess zero-DC in the primary, but it creates a large (several dB) VLF hump that is potentially problematic.
C) Add the xfmr as in A), but use one of the secondaries to drive the EQ (remove the 100r and connect the secondary across the 470k. The other secondary would be your floating record output. The idea here is that you would have the closest to a 312, with the interaction between the DOA and the xfmr.

 

[stuartpre]

Thanks 'abbey road d enfer', that makes sense to me. I like the sound of option 'C):'

"C) Add the xfmr as in A), but use one of the secondaries to drive the EQ (remove the 100r and connect the secondary across the 470k. The other secondary would be your floating record output. The idea here is that you would have the closest to a 312, with the interaction between the DOA and the xfmr."

, plus I can add a Yamaha Output transformer on the insert send (post EQ) for an alternate Tape Send with EQ in the CCT.

Hmmm, but where do I put the switch? ie 'Tape Send pre/post EQ'.... maybe re-purpose Bus 3 switch (but then I lose half my 'Yamaha buss)'...?

Hmmm, i'll be needing a direct out jack or multi-pin socket on the rear panel for Tape Sends...

Oh, and where do I mount the 2503 output xfrm? mount  in the rear with existing Yamaha transformers?

Hmm connector 2 seems to have 2 spare pins.., pins 3 and 6,  so pin 3 to send the DOA output  to 2503 xfrm and pin 6 to bring back the secondary +ve phase, while the -ve phase of that secondary just goes to earth locally?

 

[stuartpre]

I am still awaiting my CAPI parts to arrive but have begun JLM EQ ILCR mod and the first EQ complete is a solid improvement = less telephone effect, more use-able in boost, but less corrective in cut...But I like it.

Regarding the whole 2520 upgrade to the CAPI which runs on the 500 series +- 16 v , a lower voltage that the +- 24, what is the best way to drop the voltage down.

A) cut tracks on the voltage rails and wire the the +-16 volt rails from the voltage distribution drop (uses paralell Zenner's and caps which follow series resisters)

Enough current? or

B) should i attach my CAPI 2520 to a 'lil board with a similar Zener voltage drop cct?

Is A) a potential noise issue? or B) ?

 

[abbey road d enfer]

I am still awaiting my CAPI parts to arrive but have begun JLM EQ ILCR mod and the first EQ complete is a solid improvement = less telephone effect, more use-able in boost, but less corrective in cut...But I like it.

Regarding the whole 2520 upgrade to the CAPI which runs on the 500 series +- 16 v , a lower voltage that the +- 24, what is the best way to drop the voltage down.

A) cut tracks on the voltage rails and wire the the +-16 volt rails from the voltage distribution drop (uses paralell Zenner's and caps which follow series resisters)

Is it really how you generate the 16v rails? That's probably the worst option. Next is Zener drop, which requires additional decoupling caps, and last but not least, regulating via LM117/137.

 

[stuartpre]

Ok, Thanks.

 

[mitsos]

Why not modify the main PSU to put out lower DC voltages? You dont need 25v for anything.  Then you can get rid of the 16V zeners  in the eq sections.

 

[stuartpre]

Will this work? Are there any people out there who advise against this?

 

[abbey road d enfer]

Why not modify the main PSU to put out lower DC voltages? You dont need 25v for anything.

+/- 25V is used by all the IC's in the signal path. Powering the whole mixer with +/-15V would reduce the headroom by about 5dB.
Now I agree that if the OP puts a +/-15V device in the signal path, it will bring the headroom down.
Except the groups and masters will still be capable of the high headroom.
And that would bring the issue of dissipation in the PSU. It would mean doubling the regulators and heatsinks. Alternatively, having on-board regulation is just a matter of fitting two TO220 regs, two caps and 4 resistors, which is a much lesser task.
It is a pity that the OP has not deemed necessary to seek advice before ordering the DOA's.
However, I believe there is an easy mod to make the CAPI DOA's capable of operating at +/- 24V. Mailing Jeff Steiger will get an answer, or maybe it's already there http://groupdiy.com/index.php?topic=33970.0

 

[stuartpre]

What is OP? Is that me? I know you guys most likely think I am a bit lacking in knowledge for this mod, but how else does one learn? I must try and learn by my mistakes and ask a few wise people for there advise along the way. Thanks for you help so far.

I have been in contact with jsteiger in the build thread for the CAPI ACA-BO, so this thread is probably now redundant. FYI Jeff is aware of the PSU issues I am having and has not suggest modding the DOA to +- 25v. Instead I am building an additional supply +-16v using a JLM ACDC kit.  Thanks for your replies.

 

[abbey road d enfer]

What is OP? Is that me? 

Yes, Original Poster...

I know you guys most likely think I am a bit lacking in knowledge for this mod, but how else does one learn? I must try and learn by my mistakes and ask a few wise people for there advise along the way. Thanks for you help so far.

I didn't want to diss you, I just meant that this issue could have been avoided by using a different type of DOA that works on +/-24V rails.

I have been in contact with jsteiger in the build thread for the CAPI ACA-BO, so this thread is probably now redundant. FYI Jeff is aware of the PSU issues I am having and has not suggest modding the DOA to +- 25v.

Why don't you ask him specifically how to mod the 2520 to accomodate elevated rails? He should be capable of giving you an answer, it's not that complicated.

 

[stuartpre]

Thanks for the clarity.

I have gone ahead and built a separate supply for the +-16v.

The CAPI ACA-BO works just fine now. A major improvement on the Yamaha mixbuss (I still have buss 3/4 up and running so direct comparison is easy).

Perhaps the modification for the DOA may prove useful in a possible preamp swap, as previously discussed. 

I will go ahead and ask Jeff about the mod you have suggested, thanks for the input.

 

[mitsos]

Good that you got it working.

Just wanted to clarify about lowering the main dc rails. I suggested this not only to make the aca-bo work but mainly  because it would allow you to change ICs from those toshiba SIP thingies to more modern solutions.  One place to change would be the summing amps, all the old yamahas use these toshibas for summing, from the little rackmounts up to the PM2000.  Wirh the stock supply you can swap those for an OPA604, but with lower rails  you have tons more options.  That's one spot I've been planning on swapping opamps in the few old yammy's sitting around here.

Abbey is right about the lower headroom, of course, all things being equal, but there are other factors to consider.  I dont want to get you too far off topic though. 

Anyway, glad your ACA-bo worked out, they are def an improvement over most IC summing I've heard.