8 Channel, 3U NEVE 1073 completed

[sr1200]

If you can do a board with the BC's do it.  If that works, I have some spare old stock BC's I might be able to get to ya.  I only plan on building a couple more pres and I know i have more BC's than id need... (keeping some in reserve just in case )

 

[indiehouse Recording]

If you can do a board with the BC's do it.  If that works, I have some spare old stock BC's I might be able to get to ya.  I only plan on building a couple more pres and I know i have more BC's than id need... (keeping some in reserve just in case )

Sweet, thanks!

Random question, but I just want to make sure I'm not assuming something that is not correct. When there is no load on the PSU, I get a reading of 24V. When there is a load (my pre's are connected), I should still get a reading of 24V from the PS, right? Or is it a matter of each board drawing current from the PS, which in effect subtracts that from the available 24V?

 

[sr1200]

K, im lookin over one of my units now, everything looks ok so far, the only difference i see at first glance (besides the BC's) is you have the shield at the output connected to the board and i cant really tell, but is it possible the shielding is touching one of the leads on the transformer?  When you powered up, was the board IN the case?

 

[indiehouse Recording]

K, im lookin over one of my units now, everything looks ok so far, the only difference i see at first glance (besides the BC's) is you have the shield at the output connected to the board and i cant really tell, but is it possible the shielding is touching one of the leads on the transformer?  When you powered up, was the board IN the case?

Thanks for checking that out!

I don't think it is. I cut the shield end at the sleeve on the tranny side. So I have 2 leads on one end and 3 on the other.

Here's something strange. I tried hooking the board up to see if I had signal. I disconnected the JLM Go Between as well as the trim pot. Everything is bare bones.

I connected the PS and I now read 8.0V, plus I have signal! I know 8.0V is a far cry from 24V, but it is supposed to read 24V, even when a board is hooked up, right?

Also, when I shut the PS off, I get a radio station that fades in and gets real loud, then fades out. Sometimes I'll get a click or snap when I switch the PS on.

So then, I tried to repeat this for the second board. Same exact everything, no Go Between, no trim pot. For this one, my reading still reads 3.3V, and a weaker, distorted sounding signal.

I left my 48V feed going to the Go Betweens. That's not hurting anything since I'm not using it, right?

 

[sr1200]

i sent ya a PM.  Also, you should get 24v when the unit is hooked up and working properly (MAYBE a little sag if you're really loading it down, but still 24v)  If you're picking up radio you have some other issues going on (ground loop acting as an antenna or worse).

 

[desol]

Sounds like a poor ground connection between things and the losing power thing could be bad solder joints. Go over all your solder joints on one 1290 card and power supply...and all your component orientations. It's a good idea to get things going basically before adding jlm, etc.

 

[sr1200]

well this whole saga has me interested.  I had a nice lil skype session with indiehouse and we've found 2 things. 
I wanted to see if the PS was shot or not, but a dummy load was not available, so i had this idea...
1) we lowered the 48 rail down to 24 and used that rail to check the card.  That rail then failed as well once load was on it.  So I'd rule out the powersupply for now as the point of failure.
2) the 3055's he has were ebayed... im smelling the rat there.  We compared resistance ratings all over the card and for the most part got identical values.
3) we checked continuity between points and all seemed well.

My suggestion was to get new 3055's (ones from a reputable dealer lol) To me that seems to be a possible common point of failure for all 4 cards.  After that, try using the BC transistors.  If that doesn't work... indie, send me one of the cards and ill go point to point with one of mine.  I love a challenge! haha

 

[alexidoia]

It's kinda scary to see such component could be fake and still for sale.

 

[indiehouse Recording]

Yep, in hindsight I probably shouldn't have sourced Motorolla's from the bay. The seller was from the US, and had 100% perfect feedback rating. I believe he was a private seller, not a dealer, based on the low feedback score.

I did score 10 of these for a little under 8 bucks shipped, so that may have been a little on the cheap side.

I did do a little research on these before I bought them. Here is some stuff I found that may be helpful to others:

http://www.groupdiy.com/index.php?topic=42602.0
http://sound.westhost.com/counterfeit.htm
http://www.diyaudio.com/forums/parts/2848-2n3055-mj21193-mj15004-pics-detecting-counterfeits.html
http://sound.westhost.com/fake/counterfeit-p1.htm

But for now, it's off to buy some ST 2N3055's!! As much as I don't want to have bought fake transistors, I really hope this will be a fix!

 

[useme2305]

so, what's the latest news on this, indiehouse?

 

[indiehouse Recording]

so, what's the latest news on this, indiehouse?

After a Skype session with sr1200, we decided to try replacing the 2N3055 and the MPSA18's. Just waiting on parts, which should be here tomorrow.

 

[indiehouse Recording]

Well, I replaced the Motorola 2N3055 with the ST 2N3055. No change, still reading 8V on the 24V rail.

Next, I will try replacing the MPSA18's with BC184C's. I hope this works...

 

[indiehouse Recording]

Switched out the MPSA18's and installed BC184C's in one board. No change. Still reading 8.6V on the 24V rail. I am at a complete loss and am pretty disheartened...

 

[useme2305]

it's your psu man.

 

[dmp]

I'd suggest checking your psu as well, since these neve's take a lot of power.

Just connect the psu to a 2 watt resistor completely separate from the neves, the resistor between 24v and ground.
If you have a 600 ohm that will check if the psu can supply 1Amp.
If you have a 1k ohm that will check if the psu can supply 0.5 Amps

Hook it up and measure the voltage, if it is still 24 under load you are good.

 

[Harpo]

24V/600R= probably not 1A.
Sometimes a common avail.light bulb is just the needed cheap high wattage resistor ..., your might have some spare bulbs for your cars blinkers or backlights, comming in 12V/5W, 10W, up to 21W. Use two in series for 24V. From I=P/U and R=U/I, a 12V/5W bulb will draw ~0.42A and represent a 28R8 resistor. Two in series at double voltage will draw the same current. Two 12V/10W bulbs in series for a 0.83A load might tell if the size of your voltage regulators attached heatsink is sufficient.

This might help a little ...

indiehouse, from post #1505 your underside of psu shows the LM317 voltage regulators mounted upside down and in wrong pin order. This way the chips heatfins couldn't transfer generated heat to the heatsinking case when assembled and the wrong pin order now connects raw DC in to adjust pin and voltage setting resistors junction connect to regulators input instead. Bend the regulators pins upwards. You might have fixed this already.
You could ohm out, if the 2N3055 collector (case) conducts to the next soldered point on pcb thru the mounting bolt and fixing nut when you follow the pcb trace. Using a sawtooth washer between pcb and the left side transistor fixing nut from your pic to go thru the insulating coating would fix this if you don't get a low/close to zero ohm reading.

 

[indiehouse Recording]

24V/600R= probably not 1A.
Sometimes a common avail.light bulb is just the needed cheap high wattage resistor ..., your might have some spare bulbs for your cars blinkers or backlights, comming in 12V/5W, 10W, up to 21W. Use two in series for 24V. From I=P/U and R=U/I, a 12V/5W bulb will draw ~0.42A and represent a 28R8 resistor. Two in series at double voltage will draw the same current. Two 12V/10W bulbs in series for a 0.83A load might tell if the size of your voltage regulators attached heatsink is sufficient.

This might help a little ...

indiehouse, from post #1505 your underside of psu shows the LM317 voltage regulators mounted upside down and in wrong pin order. This way the chips heatfins couldn't transfer generated heat to the heatsinking case when assembled and the wrong pin order now connects raw DC in to adjust pin and voltage setting resistors junction connect to regulators input instead. Bend the regulators pins upwards. You might have fixed this already.
You could ohm out, if the 2N3055 collector (case) conducts to the next soldered point on pcb thru the mounting bolt and fixing nut when you follow the pcb trace. Using a sawtooth washer between pcb and the left side transistor fixing nut from your pic to go thru the insulating coating would fix this if you don't get a low/close to zero ohm reading.

Thank you so much for this info! I'll take a look and report back with an update.

 

[indiehouse Recording]

Do you think I damaged any other component or the regs themselves by this? Should I replace them?

 

[Harpo]

Other components should be fine, but I'd replace the vRegs. You have to desolder them anyway and when bending the vRegs pins to the opposite direction, chances are pretty high you break a chips leg. No idea if they are internally broken by now. Datasheet will only give you conditions or limits the chip was designed for, IE vRegs i/o differential must not exceed 40V... (if you have to buy new vRegs, trying to get a LM317-HV type for a little safety margin for the +48V rail might not be the worst idea, but as always YMMV).

 

[useme2305]

yup, both of your LM317s are installed the wrong way around.

shocking how no one noticed it before.

when you're done installing them correctly make sure to secure the power plant tightly to the case to ensure heat dissipation or even better attach it to a nice and big heatsink and install the whole package in a case.