Help Troubleshooting Millenia Media HV-3B PLEASE

[Ike Zimbel]

I found the same thing in the two data sheets that I had on file. I wouldn't recommend going higher than +/-20 in regular use.

 

[sonolink]

Hi guys,

I just got the OPA604s and the OPA2604s and replaced the opamps, installed sockets, reconnected the lifted diodes, hit power and OOOOOOOPS all the OPA604s went up in smoke!!!

I ckecked that they were properly oriented, blablabla and my conclusion is that this device uses OPA2604s and NE5534s since the TL071s didn't get smoked when testing.

So I'm buying some NE5534 this afternoon and will let you know

Cheers
Sono

 

[ccaudle]

my conclusion is that this device uses OPA2604s and NE5534s

That is a very odd conclusion given that OPA604 and NE5534 have the same pinout, but OPA604 can run at 4V higher supply rails. If OPA604 let the smoke out then NE5534 should do the same but a little bit faster.

 

[ccaudle]

since the TL071s didn't get smoked when testing

I just double checked and TL071 and OPA604 have the same pinout as well, so something else is going on. Put the sockets in backwards, connected the diodes incorrectly, diodes caused the first problem and did the same thing when put back, etc.
But the conclusion that NE5534 is what you need doesn't make any sense.

 

[Matt Syson]

The maximum rated voltage quoted for op amps is the Voltage that the manufacturers test for and will therefore honour the 'warranty'. Of course individual chips MIGHT be happy to run quite a few volts more than this BUT then, they might not and you can't tell by looking. there may be other more subtle things going on such as the polarity of the op amp's input transistors wo bias current flow direction, AND not all chips have the same reaction to having BOTH input pins taken close to a supply rail. The NE5534.32 chips have internal inverse parallel protection diodes between their input pins to reduce/prevent damage to the input transistors. Then again the circuit may use offset nulling of which there are several different techniques. Some designers effectively bypass the input transistors by using 'external' transistors which couple to the offset nulling pins and the internal transistors are biassed off completely. Studer (among others) used this idea so that the 'noisy' internal pair of a LM301? could be bypassed but the remaining circuitry provides the 'output' stage. As ever with op amps there are many factors to consider and as tempting as it looks, a 'simple' swap may not behave as first thought.

 

[sonolink]

I'm just saying NE5534s should work because one of the channels had one on it (the other opamps were filed)

 

[moamps]

I just got the OPA604s and the OPA2604s and replaced the opamps, installed sockets, reconnected the lifted diodes, hit power and OOOOOOOPS all the OPA604s went up in smoke!!!...

Obviously something very bad has happened, because a violent reaction means that either the power supply is incorrectly connected, or the opamp's output is overloaded, or the opamp has crashed into large oscillations. The input single opamps in this design (OP604, NE5534 or TL071) are powered by +/- 18V, and the output dual opamp OPA2604 is powered by +/- 24V.
I suggest you check what you did exactly, and try to take a few steps back and get your preamp to work with 071 again, and that you go one step at a time and test the correct operation.

 

[sonolink]

Obviously something very bad has happened, because a violent reaction means that either the power supply is incorrectly connected, or the opamp's output is overloaded, or the opamp has crashed into large oscillations. The input single opamps in this design (OP604, NE5534 or TL071) are powered by +/- 18V, and the output dual opamp OPA2604 is powered by +/- 24V.
I suggest you check what you did exactly, and try to take a few steps back and get your preamp to work with 071 again, and that you go one step at a time and test the correct operation.

I will check again but what I did was:

-take out all the opamps on both channels (since the working one was a bit funky too).

-Install DIL8 sockets on the working channel.

-Reconnect lifted legs of diodes

-Install heatsinks on opamps

-install opamps on sockets

-Switch the unit on.

 

[sonolink]

Ok, so I tried it with the NE5532 and 2x TL071s and there is no sound just some hiss.

I have checked voltages:
NE5532
Pin 4: -25
Pin 8: 24
All the other pins 0v

TL071s
P1 18v
P2 and 3 6v
P4 and 5 -18v
P6 -8v
P7 18v
P8 0v

Voltages seem fine don't they?

 

[moamps]

Those voltages look ok, but you need to take steps in the procedure we’ve already done in the last few pages of the thread. You also have the voltages you need to get at certain points, turn on your generator and oscilloscope and go. Can you send a picture of those OPA604s that burned down?

 

[sonolink]

I was afraid I had to do it all again....
Here's the pics:

 

[moamps]

Those chips seem like fakes to me. Yes, you will need to use the knowledge you have learned.

 

[sonolink]

Really?? I guess I won't be able to claim my money back, right?

 

[pucho812]

Careful with the opa2604 and 604. They are rated at bipolar 24vdc. But there was a batch or two that would fail when given +/-24vdc

 

[sonolink]

Question : should I lift the zeners again before using the scope with the generator?

Cheers
Sono

 

[Matt Syson]

The voltages you posted do not look fine. Pin 4 of the5532 should not be minus 25 volts especially as pin 8 is 24 volts.. The TOTAL between 4 and 8 should be less than 40 Volts for a 5532 or 5534.
Also the TL071, having minus 6 volts on it's output would be suspicious if it is only doing normal twin rail 'audio' stuff. If it was part of a rectifier circuit for metering or whatever then maybe OK.
Unless the design deliberately calls for asymmetric supply voltages it is usual for at least the audio rails to be 'equal' as audio signals are usually referenced to 'ground' The 'fact' you haveminus 25 volts on a supply pin is suspicious as it is a bit high, even for the OPA604 chips.

 

[sonolink]

Hi Matt,

No idea if you're right but these were the voltages on the working channel (when it worked...)

U6

Pin 1 0
Pin 2 0
Pin 3 0
Pin 4 -24
Pin 5 0
Pin 6 0
Pin 7 0
Pin 8 +24

Voltages on faulty channel were the same
Cheers
Sono

 

[Cranehazard]

Check those power supplies looks like minus 24 regulator could be blown, very easy to blow up, any small amount of positive voltage applied when turned off will cause it to blow when powered up. The 18v regulators will still read plus and minus 18v

 

[sonolink]

Ok guys, so I re-did the whole procedure tracing the signal with the scope using NE5532 in the place of OPA2604 and TL071 instead of OPA604/NE5534.
Both channels work fine.
I know I'm surprised as well....

So I replaced NE5532 with OPA2604 hoping those didn't get fried and both channels work.
I have no idea why the OPA604 went into smoke but the funny thing is that I tried replacing the TL071 with the (fried) OPA604 and smoke again. I put back the TL071s and everything fine!!

So I went down the shop and bought a few NE5534sand tested the unit with the OPA2604s and the NE5534s.
Both channels work fine!!

I have no idea why the 604s got smoked. At least I could fix the unit for my friend thanks to all of you, especially Moamps!

Thanks again!!
Cheers
Sono

 

[radardoug]

This is a very good example of why you shouldn't try and out think the original designer, and why most opamp changes are a waste of time.