thermally coupled electrolytis

[Tubetec]

Hi Angry ,
Welcome to the forum ,

 

[NoisyIndividual]

i'd like to add that the micpre uses a MAT02 .

Have you looked into the costs etc. in replacing MAT02 recently?

 

[user 133392]

I suppose I should also mention that PR&E consoles were designed for broadcast, so some of the circuitry will be a bit different from the usual pro audio stuff. Mostly subtle differences, but some will be real head scratchers. Though usually our consoles were installed in nice clean studios, sometimes they would be installed in high RF fields. We guaranteed our products would be free from RFI in any environment. So while that copper shield might not do much to attenuate RF directly, the improved CMRR at AM frequencies helped. To this day, PR&E consoles are considered by many to be the gold standard in broadcast boards. Also, Deane Jensen was our audio circuitry design consultant so some of his ideas are found in those consoles including the use of JE-990 discrete op-amps in summing amps and the generous use of Jensen transformers in some designs (not Radiomixer though, that is a solid-state preamp design).

Thanks for joining us. You make some nice stuff these days.

So it looks like that not only does the copper wrap create an electrostatic shield to the rest of the circuitry and prevent EMI ingress but a capacitive delta network is also created consisting of outer foil to outer foil capacitance (due to proximity of the cans) and then a capacitance from each outer foil to ground due to the copper wrap.

Consoles co-located with AM transmitters may be slowly disappearing but consoles located close to cell phones are not - this type of shielding is still justified. Rectification of pulsed RF remains a problem.

 

[AngryAudio]

the micpre uses a MAT02

No, it is actually an LM394. The MAT02 is a laser trimmed matched transistor pair. The LM394 is also a matched transistor pair but uses many transistors in parallel to achieve a statistical match. It is a unique device, same device used on the front end of the JE990. Though discontinued by the original manufacturer, there now a company in Latvia that makes an equivalent. I am using it in another mic preamp design now and very pleased with the results.

that PS ... that really needs some muscle.

Depending on which vintage of supply you have, it was designed to support a console much larger than the one you have. Earlier supplies were capable of powering anything from a Radiomixer-8 to a Productionmixer-28. Later, we designed a universal supply that could power any of our consoles except the STX. You will know if you have the universal supply by the breaker arrangement on the front panel. If it has separate breakers for audio and logic sections, it is an older supply. If it does not, it's a universal. The universal has a massive toroidal transformer.

 

[AngryAudio]

You make some nice stuff these days.

You also! I've looked at your products before and been very impressed. Maybe we'll have a chance to collaborate sometime.

 

[user 133392]

Thanks for the kind words.

Though discontinued by the original manufacturer, there now a company in Latvia that makes an equivalent. I am using it in another mic preamp design now and very pleased with the results.

For reader reference: https://www.alfarzpp.lv/eng/sc/AS394CH.pdf
Were those shielded cap pairs on the line inputs or mic preamp input? I'm assuming the mic preamp...

Just out of curiosity I measured the shared capacitance between two bound 47µF/100V 10mm dia x 12.5mm height caps masking-taped together.

The capacitance, measured at 10 kHz (the instrument's limit) between "-" terminals was about 4.0 - 4.5 pF. Other combinations of terminals was about a pF less suggesting, if not confirming, that the "-" electrode is the outer wrap.

When I get a chance I'll measure terminal capacitance to a foil outer wrap.

 

[nashkato]

Were those shielded cap pairs on the line inputs or mic preamp input? I'm assuming the mic preamp...

they're on both

 

[AngryAudio]

Were those shielded cap pairs on the line inputs or mic preamp input? I'm assuming the mic preamp...

Both. But counter-intuitively, it was more beneficial for the line inputs.

 

[JohnRoberts]

My first experience with high RF fields was when we sold a console installed into the main beam of an AM broadcast antenna. I flew down for a couple hour service call and ended up taking 2 days.

I still remember the station,,, 960 kHz.

JR

 

[RoadrunnerOZ]

Seems related the old rule for reducing hum/RF interference from tube stages by making sure the outer foil of “non polar” caps always went to the ground or lowest impedance side of any circuit.
In the days of AM radio it was easy for anything to become an antenna. I remember working on shielding a lot of gear and having all cabling runs in grounded metal ducting in a studio near a huge transmitter tower which used to inject into everything. The sound insulation had foil on one side all connected by bare copper wires run inside the walls and ceilings to a central main switchboard earth point.
Also applicable to any solid state builds. I’ve serviced some older gear that has had twin + rail supply electros with metal clamp shields grounded to chassis - I imagine the same thought applies there as well.

 

[Newmarket]

No, it is actually an LM394. The MAT02 is a laser trimmed matched transistor pair. The LM394 is also a matched transistor pair but uses many transistors in parallel to achieve a statistical match. It is a unique device, same device used on the front end of the JE990.

Also in SSL mic pres IIRC ?

 

[JohnRoberts]

LM394 was a nice part especially for logging applications because of the good behavior, due to highly parallel architecture. At 1nV rt/Hz it was quiet for the time, but by the late 70s I discovered some Japanese parts developed for MC head amps with even lower Rbb and lower en (2sb737 and 2sd786 they were originally from a small Japanese company that was bought by ROHM to distribute internationally). They are both now obsolete.

FWIW I made only one DOA back in the 70s/80s using a LM394 for the input stage but it was simple inverting only application so not very complicated.

JR

 

[JohnRoberts]

Seems related the old rule for reducing hum/RF interference from tube stages by making sure the outer foil of “non polar” caps always went to the ground or lowest impedance side of any circuit.
In the days of AM radio it was easy for anything to become an antenna. I remember working on shielding a lot of gear and having all cabling runs in grounded metal ducting in a studio near a huge transmitter tower which used to inject into everything. The sound insulation had foil on one side all connected by bare copper wires run inside the walls and ceilings to a central main switchboard earth point.
Also applicable to any solid state builds. I’ve serviced some older gear that has had twin + rail supply electros with metal clamp shields grounded to chassis - I imagine the same thought applies there as well.

Yup... many film caps were marked with a band/line to indicate which lead was connected to the outer wrap.

JR

 

[RoadrunnerOZ]

Though discontinued by the original manufacturer, there now a company in Latvia that makes an equivalent. I am using it in another mic preamp design now and very pleased with the results.

That would be the Alfa AS394? Is it a drop in replacement - there doesn’t seem to be a DIP version??

 

[nashkato]

If it has separate breakers for audio and logic sections, it is an older supply. If it does not, it's a universal. The universal has a massive toroidal transformer.

no breakers , but 4HE cooling rips on each side and a big "schnittbandkern" (cut band core ? ) transformer

 

[AngryAudio]

no breakers , but 4HE cooling rips on each side and a big "schnittbandkern" (cut band core ? ) transformer

The breakers appear to be toggle switches on the front panel. If there are separate "switches" for logic and audio supply, it is an older supply. It sounds like you have one of the newer universal supplies. It is a better design actually.

RoadrunnerOZ:​

That would be the Alfa AS394? Is it a drop in replacement - there doesn’t seem to be a DIP version??

As far as I can tell, yes. I've looked over the data sheets and it does appear to be the same. I'm using the SOIC-8 version but they also have through-hole can which is exactly what we used at PR&E.

 

[RoadrunnerOZ]

The breakers appear to be toggle switches on the front panel. If there are separate "switches" for logic and audio supply, it is an older supply. It sounds like you have one of the newer universal supplies. It is a better design actually.



As far as I can tell, yes. I've looked over the data sheets and it does appear to be the same. I'm using the SOIC-8 version but they also have through-hole can which is exactly what we used at PR&E.

Ok cool thanks - I have a synth in which someone has asked me to complete their unfinished build and it has 4 of these which were not wired into the boards fully and I fear they may have blown

 

[user 133392]

That would be the Alfa AS394? Is it a drop in replacement - there doesn’t seem to be a DIP version??

I've seen quite a few LM394H TO5-8 lead-formed into a DIP pattern and inserted into sockets in my former life. Harrison consoles come to mind.

Also in SSL mic pres IIRC ?

Yes, at least in the 4K 6K active input preamps and IICR the 9K. https://www.ka-electronics.com/images/SSL/ssl_82E149.pdf

Both. But counter-intuitively, it was more beneficial for the line inputs.

Would that be because line input tie lines could potentially be longer than mic inputs in a typical broadcast plant facility therefore be better antennas?
An input on a remote line selector going to a TOC could be really long.
Or did the mic inputs just have better CM rejection at RF...

 

[AngryAudio]

Would that be because line input tie lines could potentially be longer than mic inputs in a typical broadcast plant facility therefore be better antennas?
An input on a remote line selector going to a TOC could be really long.
Or did the mic inputs just have better CM rejection at RF...

Mics by definition, are local to the studio. Lines could be very long as you say, but potentially even longer. Back in those days, you might bring in signals via analog (balanced) line from other buildings on campus or... the very worst-case, a 15kHz telco pair which might be a mile or more.

 

[RoadrunnerOZ]

I've seen quite a few LM394H TO5-8 lead-formed into a DIP pattern and inserted into sockets in my former life. Harrison consoles come to mind.

Careful work with a pair of pliers there. I have bridged SMD into IC sockets to make a DIP chip in the days before you could buy adapters as well. The bad thing about the multi-pin TO5’s is that when you get them they are often loose in a pack with the legs all twisted like a jellyfish