Some API pres finished

[John Hardy]

The original 990 op-amps were potted with a clear epoxy encapsulant and silicone molds.

When the "A" and "C" versions were introduced around 1988 or so, I began using a black-anodized aluminum shell with silicone encapsulant. It is a better package from a performance standpoint. Not as fun to look at, but the silicone encapsulant is soft and does not apply as much stress on the components as the hard epoxy as things heat up and cool down, therefore expand and contract. The aluminum shell does a better job of heat-sinking the two power transistors which create most of the heat. Those two transistors were moved to the edge of the p.c. board so they are in direct contact with the wall of the shell. The aluminum shell also dissipates the heat better than the epoxy.

I usually bring an unpotted 990C to the AES conventions so folks can see what's inside. Thanks.

John Hardy
The John Hardy Co.
www.johnhardyco.com

 

[guavatone]

Thanks John, they look very cool that way. I would think there would be a risk of the components heating up and melt that clear material since all the components are surrounded by it. It would be cool to experiment with some Auto RTV and a square mold.

 

[David Kulka]

Volker, that's beautiful work, really! I'm extremely impressed. That 312 card you bought from me looks much better in your custom rack than it ever looked in my parts box. (But please, recap it, ok? It's begging for some new capacitors, to match the quality of the parts that you wrapped around it.)

 

[John Hardy]

Thanks John, they look very cool that way. I would think there would be a risk of the components heating up and melt that clear material since all the components are surrounded by it.

Under normal operating conditions it is not a problem. But if the 990 failed in a manner that consumed large amounts of current such as an output transistor short-circuit from collector to emitter, the transistor would seriously overheat, then the epoxy would get too hot and start to break down. It had an unmistakable smell!

In addition to causing much less physical stress on components, the silicone encapsulant is also able to operate at much higher temperatures than the clear epoxy (and most other typical epoxies).

John Hardy
The John Hardy Co.
www.johnhardyco.com

 

[[silent:arts]]

[quote author="David Kulka"]Volker, that's beautiful work, really![/quote]
thanks David.
your card isn't racked yet, but it will fit next to the BA312 card (recaped of course). I'm still searching for those switches ("Schlafaugenschalter")to fit the look.

[quote author="John Hardy"]Under normal operating conditions it is not a problem. But if the 990 failed in a manner that consumed large amounts of current such as an output transistor short-circuit from collector to emitter, the transistor would seriously overheat, then the epoxy would get too hot and start to break down. It had an unmistakable smell![/quote]
yes, the smell of the melted epoxy ...
unfortunately one melted down after using the DI input for a few hours.
I'm still searching what caused this fault, since I want to be 100% safe before I put another (clear epoxy) 990 in it.
fact is (testing with a 990C): using the DI Input heats up the 990 after a very short time (much more then using the Mic Input for hours).
could it be that the 990 can't handle the high impedance input ?

 

[John Hardy]

using the DI Input heats up the 990 after a very short time (much more then using the Mic Input for hours).
could it be that the 990 can't handle the high impedance input ?

As Deane Jensen wrote in his 990 paper (published by the AES), the 990 can be used to improve an "Input stage for any application where the source impedance is 2500 ohms or less." (and various other things, such as line drivers, summing amps, etc.)

I have done very little testing with high source impedances. Perhaps you could measure the current draw and monitor the output with a scope to see what is going on in the DI application. The 990 should be consuming approximately 25 mA. If so, it can only get as hot as 25 mA will allow it to. I'm not familiar with the circuit, so I don't know the details and possible adverse effects.

John Hardy
The John Hardy Co.
www.johnhardyco.com

 

[[silent:arts]]

[quote author="John Hardy"]As Deane Jensen wrote in his 990 paper (published by the AES), the 990 can be used to improve an "Input stage for any application where the source impedance is 2500 ohms or less." (and various other things, such as line drivers, summing amps, etc.)[/quote]
John,
thank you very much for your answer :thumb:
the DI Input has much more Z than 2500 ohms (1Meg), my consideration was thus correct.
should have read (and tested) more before putting them in.
but hey, my fault sounded great (until it smelled bad) :green:
I will leave the DI Input out in my next two channels, and have to find a solution for the DI Input in the existing two channels.
thanks again for your support (and the great sounding opamps) :grin:

 

[Reptil]

:shock: :grin:
wauw

 

[Sun King]

Hi,  would absolutely LOVE to see these pictures, but alas, I can't see them.  Are they stored anywhere?

SK