Great River MP-2 , info, schematics & discussions

[okgb]

I try to keep the first page with all critical info , but
you may have to read through the thread .
the great river doc.s should be in the grpmail account
here

You can use any 1:4 or  1:5  ratio input transformer
the Lundals / Jensen/ Cinimag  are proven performers

Output transformer is optional  , not needed but usefull

The preamp cards have regulators for the supply voltage
so you can feed them " raw " d.c. from a very simple supply

The switches just close a contact point you can use a common ground
from the board for the two of them , spst unless you want to switch an
led for the 48v supply

 

[pacemaker]

Hi greg,
thanks for your reply,
i've read and read the thread ,
but couldn't find answers about the switches.
I will look in the gmail account to see if there's additional in fo there,

as for the edcors,
dan posted something about the MX5cs
but not about the MXL5cs,wich seems to be a new product
But as everybody seems to use the cinemag or lundahl,
i think i will use one of those irons,
For PSU, does using the allready regulated boards form peter
is overkill ?
if using them, is it ok to supply the boards with regulated 24V and 48v ?
Regards,
Francois
Regards,
Francois

 

[Dan Kennedy]

Hi Sriyan,

The voltages don't make sense. The outputs of the regulators should be +14 and -14.

The reference voltages should be about +7 and -7.

The comparators should trip about 3dB down from clipping, about +26dBu.

Measure this at the output resistor (10 ohm, 1W) on the audio board. There is
a 9K09 resistor in series with the input, which forms a divider with the 10K tie-down
resistor I talked about in the previous response. I had forgotten about the 9K09 resistor,
so in reality, the 10K's are essential.

Francois,

I don't recommend using outboard regulation, unless it's set to give more than + and - 27 and 52
volts. This allows the onboard regulators to operate properly. If they are fed less they will not be
working correctly, making a sloppy supply out of perfectly good ones.

I don't know what Peter's supply is capable of, but you might be able to use it's transformer and rectifier
and filtering, just bypassing the regulators, feeding the raw dc to the MP-2  boards.

The Edcors look good electrically, but lack shielding, so are not really suitable unless you have access to mu-metal to
make magnetic screens. At this level of performance you aren't doing yourself any favors using anything
but the best transformers.

I'm not sure what the question about the switches is, they simply pull in the relays that do the functions on the boards.

 

[speiris]

Hi Sriyan,

The voltages don't make sense. The outputs of the regulators should be +14 and -14.

The reference voltages should be about +7 and -7....

Hi Dan,

Reference voltages are now correct.  I had been careless in hooking up the LM337 and had done so incorrectly.

By the way, which resistor is the feedback resistor?

Thanks
Sriyan.

 

[Dan Kennedy]

Feedback resistor for what? There are none in the clipping circuit.

If it's for the amplifier, it's a 2K74 ohm resistor connected between
the four 20 ohm emitter resistors of the output stage and the gate
of one of the FETs.  R28 in my drawing.

 

[pacemaker]

Thanks for your reply Dan.

about the edcor,
You talked about the MX5cs :
http://www.edcorusa.com/Products/ShowProduct.aspx?ID=343
wich are not shielded.

but i saw their new MXL5c :
http://www.edcorusa.com/Products/ShowProduct.aspx?ID=352
wich is shielded.

As i wrote earlier,
i will use cinemag or lundahl,
But i thought it could be helpfull for future builders.
And these trannies seems a lot bigger than the MX5cs
maybe they're better ;-)

i think i have understood the switches
the impedance and polarity switches are sharing the 24v
wich is pin 3 of J6 right ?
regards,
Francois

 

[speiris]

Feedback resistor for what? There are none in the clipping circuit.

If it's for the amplifier, it's a 2K74 ohm resistor connected between
the four 20 ohm emitter resistors of the output stage and the gate
of one of the FETs.  R28 in my drawing.

Hi Dan,

It was for the amplifier.  I was reading another thread where someone wanted to place a pot on the output. 
You mentioned using a pot and putting a 1.5K resistor in place of the feedback resistor to be able to get a few dB up and down from the gain value set by the rotary switch.

Is that a pot placed directly on the output and a 1.5K in place of R28?

Not too intent on implementing this but I'm just interested in learning how it works.

This is my first electronics project and I've learned quite a bit already.

Thanks for providing this project and again for your advice and help.

regards
Sriyan

 

[Dan Kennedy]

Hi again guys,

The Edcor is still not magnetically shielded, the copper tape is good for
electrostatic interference, but does nothing for magnetic (hum) fields.

At microphone levels, in a metal chassis, the magnetic fields are the bigger problem, as
the chassis will shield a lot of electrostatic noise. That's why I mentioned mu-metal, it
is a very effective material for magnetic shielding.

To adjust the output level of the amp, the feedback resistor needs to vary, so the way
I would do it is to put the 1.5K resistor in series with the pot, which should be wired as
a rheostat. This network then replaces R28. The leads should be kept pretty short, and
away from the input wiring as much as possible.

I don't really recommend it, by the way.

 

[riggler]

Oh man, I screwed something up....

???

First channel is ok, I needed 980R to trim R46 for 0 DC offset.

BUT,

Getting 19V DC on the output of my 2nd channel!!

I'm going to pull the board and look for shorts, etc.

On this board, when soldering components, I accidentally had Q3 and Q4 swapped. I had a heck of a time desoldering and swapping them. I wondered even then if these got overheated during that process, could that cause this?

EDIT: I double checked for shorts and I think I'm good there. I found that I have continuity over pins 2 and 3 of Q3 (2N4403) in-circuit. Is this bad? Well, according to the schematic pins 2 and 3 are tied via a small jumper so I guess that's okay...

EDIT 2: Checking voltages around the boards (I do have one good channel to compare). Negative meter lead clamped to the ground point, probing with positive lead. My "good" board shows 1.25V at the outside lead of the 38.3R resistor next to Q14 and on the bad board it's 20.8V!

 

[okgb]

inject a signal and see where it stops [ osciliscope ]
if the soldering and values look right at that point
then subbing a couple of spares seems a reasonable next step
good luck , but you got one right so bravo as well

hope it could be as simple as a cap , you have some place to start
you could replace things one at a time but it's allot of work
and you don't want to fry parts , better to work with logic
but basicaly start in signal chain order

 

[riggler]

OK Greg and all,
A little update. I don't have an oscilloscope (I know.......), but I did more meter probing and found the following:

C21 - the 75pF cap, on the bad board is showing about 20V on both sides! On the good board, it shows about 20 volts on one side and 1.3V on the other.

Here's a table of some other measurements...

Point_________	BAD board	GOOD board
Q4 Pin 3	20.7V	20.7V
Q8 Pin 3	20.8V	1.3V
Q11 Pin 1	-23V	-23V
Q13 Pin 1	18.2	-1.3
Q13 Pin 3	20.8	1.3
Q13 Pin 2	18.9	-.5
Q14 Pin 1	20.25	.7
Q14 Pin 2	20.8	1.3
Q14 Pin 3	23.11	23

 

[Dan Kennedy]

I would bet that if you replaced Q2, Q3, and Q4 it'll be ok.

If you want to send the board in I'll fix it.

You can PM me if so.

 

[okgb]

What ?
you never had to fix one cause they never broke right ?

Dam good of you Dan

 

[riggler]

I would bet that if you replaced Q2, Q3, and Q4 it'll be ok.

If you want to send the board in I'll fix it.

You can PM me if so.

Unreal! But I will get this fixed -- at least really try!! PTown is graciously going to send me some replacement transistors and the Mica cap and I will give it a whirl.

God bless you Dan! You are an inspiration to us all man!!

UPDATE:

OK, so I don't know why I didn't check voltages around the regulators before, but this is interesting:

The numbers work like this:

BAD CARD (GOOD CARD)

The 317's and 337's match almost perfectly. It's the 182 and 172 that's weird. Could bad Q2,Q3,Q4 make these voltages appear here?

 

[Dan Kennedy]

Yep.

It looks like most of the amplifier is "off".

Very little current flow, and all hanging on the positive rail.

This is a direct coupled multi-stage amp, and each succeeding stage is
dependent on the previous stage to be correct to operate.

It can be tricky to find what died, because of this, and voltages don't
necessarily mean a lot since there isn't current flowing to put them in
the proper places.

This is where proper de-soldering equipment is very helpful, you can
remove suspected devices from the board and test independently.

 

[W DeMarco]

Ive been searching the forum for pcb's for this project.  I heard great things about Dans equipment and was hoping to build a pair.  Are there no longer pcbs available?

 

[riggler]

You can get kits from PTownKid at diypartssupply.com!

 

[W DeMarco]

Cool, Found it!  I'm assuming the kit is for 2 channels as it is called the MP2.  Is this so?

 

[Ptownkid]

no, the kit is one channel of mp-2

 

[riggler]

OK I got the old Q2,Q3,Q4 out of the board. PTown is graciously sending me some replacements.

I have another sinking feeling though... Was I not careful enough de-soldering? Some of the nice shiny ring circle pads are gone. Iron temp was at 625, but I hope I didn't ruin any of the vias. Does this look okay to you guys? I was able to get the old parts out pretty easily, but remember, these have been de-soldered once before...

Pics here:

Sorry about the size... I'm going to put a lead in there and see if I can get continuity to the next via on the PCB at least per the schematic!!

AHA!!! So I tested the removed Q2,Q3,Q4 on my multimeter and they all look good! (testing resistance from base to emitter, base to collector, and then collector to emitter), BUT I went and looked at Q12 because I noticed very slight dark splotchiness, almost like flux, at the leads on top. Putting my meter across collector to emitter shows 7k!!! On my good board, shows as open!! These were both measured in circuit...