Amek Angela - Master Module - Replacing Tantalum with Electrolytic?

[smilan]

Hi,
I had a problem with bad solder in pin 1 of the M2 connector.
While reconnecting the module to the console I've connected M2 in a wrong way and as I was turning the PSU on C11 explode like a rocket...
Since I don't have in hand a 1uF/35v tantalum cap I've replaced C11 with 1uF/50V electrolytic.
There's a special reason for using tantalum caps on C11 and C12 or it's OK to replacing them with good quality electrolytics?

*Full size image
http://i66.tinypic.com/k1qwl4.jpg

 

[JohnRoberts]

Tantalum cas were popular for lower ESR than aluminum, but has a bad habit of spectacular failure modes.

Aluminum with maybe a 0.1 ceramic disc in parallel, or maybe use a modern low ESR aluminum..

JR

 

[smilan]

Thanks JohnRoberts,
Just from curiosity, why the engineers who designed this console decided to put in this place in the circuit tantalum caps instead of aluminum?

 

[ruffrecords]

Thanks JohnRoberts,
Just from curiosity, why the engineers who designed this console decided to put in this place in the circuit tantalum caps instead of aluminum?

Probably because when it was designed, Tantalum were the only available high value small size capacitors.

Cheers

Ian

 

[RuudNL]

I know that Geoff Tanner  (ex Neve) always warns NOT to replace tantalium capacitors, because "it changes the sound".
No idea if this is still true or not. ("Those tants aren't there without a reason", he says.)

 

[JohnRoberts]

Probably because when it was designed, Tantalum were the only available high value small size capacitors.

Cheers

Ian

+1 what Ian said...  we now have better/cheaper options with low Z aluminum

JR

 

[mjrippe]

I know that Geoff Tanner  (ex Neve) always warns NOT to replace tantalium capacitors, because "it changes the sound".
No idea if this is still true or not. ("Those tants aren't there without a reason", he says.)

In the audio path this is true.  However some Neve boards (notoriously the 54 series) used tants for decoupling the power rails.  These should ALWAYS be replaced with electrolytics.  Trust me, I have smelled the smoke.

 

[zamproject]

Hi

I know that Geoff Tanner  (ex Neve) always warns NOT to replace tantalium capacitors, because "it changes the sound".
No idea if this is still true or not. ("Those tants aren't there without a reason", he says.)

Do tantalum capacitor still perform better regarding capacitance VS freq ? (as ESR vs freq ?)
IIRC I see some plot comparing AL and TAN, the capacitance de-rating is way higher with AL, and it start in the audio band
Won't this have an impact with phase shift VS freq in our audio coupling situation ?

Best
Zam

 

[JohnRoberts]

Hi

Do tantalum capacitor still perform better regarding capacitance VS freq ? (as ESR vs freq ?)
IIRC I see some plot comparing AL and TAN, the capacitance de-rating is way higher with AL, and it start in the audio band
Won't this have an impact with phase shift VS freq in our audio coupling situation ?

Best
Zam

Tantalum caps were good for what they were several decades ago... now we have better options.

even I'm getting tired of repeating this story, but I only used tantalum caps on purpose in one recent (cough) design (back in the 80s). I was making a kit with dynamics processing... the encoder I was trying to match used a tantalum cap in it's side chain attack/release, so I likewise used a tantalum in the decoder attack/release. Tantalum is known for lousy DA (dielectric absorption), so I wanted to be similarly bad to track better.

If the tantalum dielectric is audible in a typical audio path, there may be something marginal (wrong) with the design.

Relax, don't worry, use a low Z aluminum.

JR

 

[Whoops]

Thanks JohnRoberts,
Just from curiosity, why the engineers who designed this console decided to put in this place in the circuit tantalum caps instead of aluminum?

"Tantalum cas were popular for lower ESR than aluminum" JR

"Probably because when it was designed, Tantalum were the only available high value small size capacitors."  Ian

You're perfectly fine for that aplication with modern Low ESR Electrolytics

 

[bxt403]

I know that Geoff Tanner  (ex Neve) always warns NOT to replace tantalium capacitors, because "it changes the sound".
No idea if this is still true or not. ("Those tants aren't there without a reason", he says.)

I got into refurbishing synthesizers a few years ago and that's where I've learned replacing tantalums blindly can cause more problems and where you can usually hear a drastic difference.  Generally: I'd replace an old tantalum with a new one.  The only exception is for the power supply/rails (where I use a low ESR, long-life electrolytic).

On a Korg M1R-EX: I replaced a tantalum with a Nichcon PW and got hanging notes.  The problem was gone once I put the original cap back in.

I'm currently changing out the tantalums on an old Studer mixer from the 80s and have decided to replace the old tantalums with new ones.  Most of them are in the signal path so using anything other than tantalum WILL change the sound (in that it's either going to improve or worsen the sound of the mixer to the extent that it might not sound like a Studer). 

Also look at pictures of any Neve 1073/1272 clone and notice they all still use a few tantalum capacitors, as it's part of the  sound of the circuit. 

As such: I'd look at the schematic and be careful about replacing the ones in the signal path.  At the same time: I'd want to swap out a tantalum with a modern electrolytic if it's connected to a power rail.

 

[jensenmann]

In this particular spot the tants decouple the rails of the subboard from the main PCB. To make that more effective you can increase both caps as you like. A rule of thumb from decades ago was to use electrolytics with 10x capacity of the tant you´re replacing. Adding a small 0,01-0,1uF ceramic cap parallel helps, as mentioned above.

 

[abbey road d enfer]

In the audio path this is true. 

Decoupling capacitors are in the signal path.

 

[gar381]

These are rail caps... replace them with Panasonic FR series.

GARY

 

[fazer]

I’ve read in wireless to not use decoupling to ground but from v+ to v- instead.  It avoids noise into the audio ground but all my old gear always has traditional bypass to ground.  Would be interesting to know more about this.  Voltage rating would double.

 

[Monte McGuire]

One problem with replacing Ta power supply capacitors is that the 79xx regulators and a few others require their odd impedance for stability. A modern low impedance electrolytic or ceramic capacitor across the output will have too little ESR and will make some of these regulators unstable.

So, while it's a nice safety feature to remove Ta from power rails, you absolutely must examine all of the regulators to see that they are not oscillating. Their oscillation looks like an odd trapezoidal waveform, not the typical sine wave of an unstable amplifier. To fix it, you have to check the datasheet for recommendations or "try and see" with different aluminum electrolytics until the regulator is stable again.

Always check the datasheet, and then check the finished circuit with an oscilloscope. Back in the 70s, it seemed like a good idea to use the impedance of Ta to stabilize a regulator, rather than requiring extra compensation components, but times change. These days, few people need negative regulators, so there aren't many physically compatible replacements that don't require Ta either. However you address it, blind replacement of Ta with Al does not always work. Check the results!

 

[JohnRoberts]

Decoupling capacitors are in the signal path.

Arguably everything is "in" the audio path, but using superior capacitors can make their audible artifacts lesser, a good thing IMO.

Of course people can use whatever they want......  On paper a new tantalum "should" be more like an old tantalum, but they probably improved over the years too.... Hopefully the new ones don't make such spectacular failures.

JR

 

[abbey road d enfer]

I’ve read in wireless to not use decoupling to ground but from v+ to v- instead.  It avoids noise into the audio ground but all my old gear always has traditional bypass to ground.  Would be interesting to know more about this.  Voltage rating would double.

It's just ridiculous. Anyone who understands Kirchoff and Thevenin knows that current that is delivered to a load must be returned to the reference point, by the shortest route.
"Noise" is not strictly noise in that case, it's an image of the audio current that circulates in the circuit. Rather than seeing the sum of currents circulating through the PSU, it's better neutralizing them at the source.
What "wireless"?

 

[moamps]

Douglas Self,
https://www.eetimes.com/document.asp?doc_id=1278963&page_number=2

"The 5532 and 5534 type op-amps require adequate supply decoupling if they are to remain stable, otherwise they appear to be subject to some sort of internal oscillation that degrades linearity without being visible on a normal oscilloscope. The essential requirement is that the positive and negative rails should be decoupled with a 100 nF capacitor between them, at a distance of not more than a few millimeters from the op-amp; normally one such capacitor is fitted per package as close to it as possible.

It is not necessary, and often not desirable, to have two capacitors going to ground; every capacitor between a supply rail and ground carries the risk of injecting rail noise into the ground."

Wireless World magazine maybe.

 

[abbey road d enfer]

Douglas Self,
https://www.eetimes.com/document.asp?doc_id=1278963&page_number=2

"The 5532 and 5534 type op-amps require adequate supply decoupling if they are to remain stable, otherwise they appear to be subject to some sort of internal oscillation that degrades linearity without being visible on a normal oscilloscope. The essential requirement is that the positive and negative rails should be decoupled with a 100 nF capacitor between them, at a distance of not more than a few millimeters from the op-amp; normally one such capacitor is fitted per package as close to it as possible.

It is not necessary, and often not desirable, to have two capacitors going to ground; every capacitor between a supply rail and ground carries the risk of injecting rail noise into the ground."

Wireless World magazine maybe.

Repeating the same mistake does not make it truer. Stability of the opamp is just one criterion. Making sure that load currents are properly retiurned where they belong is an equally important parameter.
Dealing with return currents in a large-scale mixer is very different than optimizing an opamp eval board.