How fast a 2520 can be destroyed ?

[rodabod]

Looks ok to me.

My first point though is; can we measure the input at the inverting pin when configured as an inverting opamp? The pin sits at a virtual earth.

You will need an input resistor to set the gain for this amp if you want to use it in this state by the way.


Roddy

 

[rodabod]

[quote author="SSLtech"]There's no DC reference connection to pin 3... connect a resistor to ground and repeat the test...[/quote]

Oh. Yes!

 

[SSLtech]

[quote author="rodabod"]My first point though is; can we measure the input at the inverting pin when configured as an inverting opamp? The pin sits at a virtual earth. [/quote]

Bzzzzt!

-Sorry Roddy.

The answer on the card says that the op-amp's output will swing until the input nodes are at the same potential... injecting (for random example) 7V at the non-inverting node will swing the output until either the inverting node sees the same 7V, or until the output clips... whichever happens sooner.

-My producer is telling me that we can't accept that as an answer because it's only true if the non-inverting pin has a connection to ground.... -sorry.

-But nobody goes away empty-handed, because you still have that marvelous 'teasmade', which nobody can take away from you... thanks for playing, let's have a big round of applause...

:green:

Seriously though, take note... that information unlocks the key to many things when it comes to 'seeing' how Op-Amp circuits work.

Keith

 

[vertiges]

OK...

I've got a lot of problems with the server of Prodigy. I've trying to access the forum for more than 1 hour in vain. Now it's working. :?

Keith,

I did what you suggested :

I used a 2k resistor between the IN+ and the ground.

With the 5534 :

OUT : 8.5mV
IN+ : 0.9 mV
IN- : 0.6 mV
V+ : 16.34 Volts
V- : -16.36 Volts

With the 2520 :

OUT : 1.1mV
IN + : 0
IN - : -0.5 mV
V+ : 16.34 Volts
V- : -16.36 Volts

Here is my schematic :
(there is not v+ and v- pictured, but in the real live they are present)

I'm noticing there is no ground reference for the IN+ in this circuit.
Am I wrong ?

eD

Guys, again, thank you very much !!!

eD)))

PS : it's not a 4 years old kid who have made this drawing, it's me ! :green:

 

[AnalogPackrat]

Yes, your drawing has no ground reference on the op-amp input. But it should!

API-312 Schematic

Note that "COM" is ground on the above schematic.

 

[vertiges]

On this schema from 3nity, is there any ground ref for the IN + ?
I'm confused... :roll:

EDITED :

[quote author="AnalogPackrat"]Yes, your drawing has no ground reference on the op-amp input. But it should!

API-312 Schematic

Note that "COM" is ground on the above schematic.[/quote]

So the following schematic is wrong ? :shock:

 

[vertiges]

I've added a link to ground in the following schematic.
Is that right ? :?:

eD)))

EDITED :

Could that link to ground could be elsewhere ?

BTW, I don't use RC and CC. The 75101AC doen't need it.

 

[AnalogPackrat]

Yes. Otherwise your + input sees no signal since the input transformer secondary is fully floating.

 

[rodabod]

Vertiges, yes tying the lower leg of the secondary to ground gives you your ground reference.

[quote author="SSLtech"]
The answer on the card says that the op-amp's output will swing until the input nodes are at the same potential... injecting (for random example) 7V at the non-inverting node will swing the output until either the inverting node sees the same 7V, or until the output clips... whichever happens sooner.
[/quote]

Ah, thanks Keef. It's all (well some of it anyway) coming back. The reason I suggested the virtual earth was because I had an issue the other day measuring the input into an inverting opamp - I was measuring nil at the input pin and needed to measure prior to the input resistor to get a reading... which is where my question came about!

 

[vertiges]

Thank you rodabod !!!

Just by curiosity, if I tie the upper leg to ground, I've got a "mute" ?
I mean IN + will be always be null ?

Ah well, a missing wire to ground can cause a lot of damage !

I'm gonna try to resolder the output TX, add the wire to ground, put bakc the 2520 in place, and re-test again. :thumb:

eD

 

[SSLtech]

[quote author="rodabod"] The reason I suggested the virtual earth was because I had an issue the other day measuring the input into an inverting opamp - I was measuring nil at the input pin and needed to measure prior to the input resistor to get a reading... which is where my question came about![/quote]
Yes, basically to make a virtual earth amplifier all you need to do is connect the non-inverting input to earth, and the output will move heaven and earth trying to drive the non-inverting input to be equal to ground... i.e. "virtually" ground... hence the name.

Vertiges, -sounds like you have found the problem. -Yes, as I mentioned earlier, an un-referred input will cause the output to slew dramaticlaly to one rail or another.

Yes, the schematic you refer to is missing the ground, and therefore wrong until corrected.

You may have done something nasty to the core of the output transformer... I'll let others direct you as to what best to do.

Adding a 'mute' link to the 'upper' end of the inout TX secondary will indeed mute the circuit, but it will also short the secondary of the input TX. -If you do this with a high signal present... -well... just don't do it, that's all. :twisted:

Keith

 

[vertiges]

Keith,

I'm sorry I'm trying to follow your discussion with rodabod about "virtual earth", but I must be too green in electronic ! :wink:

[quote author="SSLtech"]
Vertiges, -sounds like you have found the problem. -Yes, as I mentioned earlier, an un-referred input will cause the output to slew dramaticlaly to one rail or another.[/quote]

I've learn so much in the last 48 hours...

[quote author="SSLtech"]
Adding a 'mute' link to the 'upper' end of the inout TX secondary will indeed mute the circuit, but it will also short the secondary of the input TX. -If you do this with a high signal present... -well... just don't do it, that's all. :twisted: [/quote]

Ok, I won't. I've to confess, I was thinking about doing it because I'd like to have a mute on my preamp. But I will find another way !

[quote author="SSLtech"]
You may have done something nasty to the core of the output transformer... I'll let others direct you as to what best to do.
[/quote]

:shock: I hope not !!!

I've put back the output TX and a 2520 in the circuit, I've added the wire to ground, and I've tested my preamp with a SM57 and headphones.

THE GOOD NEWS : It works quite well, I'm able to get a lot of gain ! The sound is not distorted, but I just have a lot of "static noise", but my chassis is not grounded yet and I'm plugged in a cheap little Alesis/wallmart_PSU mixer ! Just switching off the soldering iron reduce the noise by at least 20 dB !

What would be the symptoms if the output TX is damaged ? I mean in the point of view of the sound ?

THE BAD NEWS : one of my 2520 is dead !

eD

 

[vertiges]

I've got two questions :

On this following schematic :

1) I've added two arrows. Is this how the current flows in that part of the circuit ?
2) I'm wondering about the value of RGAIN (or R2 on the other schematic.)
This resistor limits the maximum gain of the 2520 rigtht ?
The value of the rev log pot is supposed to be 22k. As I use a 20k, am I not supposed to put a 2k2 instead of the 200 Ohm ?
Is it possible to burn a 2520 if this resistor is too small ?

Actually, if I count well... there are 4 questions !!! :green:

Thanks you,

eD)))

 

[AnalogPackrat]

Your output transformer core may be partially magnetized, but you can correct this by putting a relatively high level, low frequency signal through it for some time. Once you get your pre working, just burn it in by putting a 20-40Hz signal through it with the gain up enough that the output is high, but not distorted. I believe CJ had a post or two about this in the past. Try a search.

A P

 

[vertiges]

Thank you AP !

If there are some posts about that, I will find them !!!!! :twisted:

eD)))

 

[AnalogPackrat]

It was PRR, not CJ. Here's the post:

LINK

 

[SSLtech]

...And a load on the transformer, of course. Otherwise no luvverly AC current flows.

Regarding current flow, don't forget that this is AC that we're talking about... so it will flow that way one half cycle, then the precise reverse for the next half-cycle... but basically yes, your drawing makes sense.

Remember, PRR is talking about the slight milliamp trickle from an OHM-METER magnetising a transformer core... we've been banging a few AMPERES into this one...

The result -correct me if I'm wrong- of a magnetized core is an asymmetric wave transfer. -You would experience a softly distorted waveform, probably altering significantly with level.

Keith

 

[vertiges]

Yes it's AC, you're right.

I've just finished to read the post from PRR... :shock: It's dangerous to use a ohm-meter on a transformer ????? :shock: :shock: :shock:

I've been obliged to check the wiring of the input TX with my ohm-meter !!! :roll:

If there is distorsion I will hear them in my control room. (My ears are very sensitive to distortion and phase)

Thank you Keith,

eD)))

 

[AnalogPackrat]

Not multiple amperes...maybe one. IIRC, a real 2503 has something like 20 or 30 ohm DCR per winding which is the limiting case for DC. So:

I = V / R
I = 16.5 / 20 = 0.825A

<edit> search reveals DCR is more like 6-7 ohms...

I = 16.5 / 6 = 2.75A

I stand corrected!

I think Keef is right about the effect of a magnetized core. If a scope is available, you could monitor the output waveform and see if the shape changes as you bang away at it with the low frequency signal...

Oh, and yeah, the load. Put a ~1k @1W resistor across the output.

A P

 

[SSLtech]

[quote author="vertiges"]If there is distorsion I will hear them in my control room. (My ears are very sensitive to distortion and phase)[/quote]
I hear what you're saying, but beware...

This is not like any distortion that you've commonly heard. It's not clipping, it's not crossover distortion, and it's almost certainly lopsided. The presence of 'half a good waveform' tricks the ear... -A sort of "trompe l'oreille"!

A 2-channel scope set to subtract A from B will reveal any distortions. -If you can't dial down to a perfectly flat line, there's distortion.

Keith