2 Crown DC-300A's

[RoadrunnerOZ]

I was working at Peavey when we decided to stop putting all ICs in sockets. Indeed sockets save time and money performing repairs but analysis revealed that ICs got better so failed less frequently, and most failures occurred in input or output sockets where failure stresses likely came from the outside world.

We had machines that inserted both the sockets and the ICs so labor was not a big cost, and sockets when you buy them by the truckload are not very expensive, but they were not a zero cost, and a secondary issue came come from sockets that develop bad connections over time. IIRC we kept sockets under the IC in inputs/outputs.

Of course changing to SMD technology made sockets moot...

JR

The failure of IC’s in the early days has now been replaced by crumbling IC sockets from plastic degeneration for a lot of vintage gear - but at the time these pieces of gear were current the failure rate of IC’s was high so servicing was made easier. One of the problems I’ve found has been the random breakdown of contact between IC pins and the socket jaws from oxidation/tarnish or socket cracking, causing intermittent operation and sometimes HF noise.

 

[RoadrunnerOZ]

At Peavey we had to honor many loudspeaker warranty claims for speaker failures that were caused by other manufacturer's amps. I do not recall any (many?) that were related to DC300 specifically, but being DC coupled the DC300s were notorious for that. The biggest warranty suck hole that I recall was caused by the current limiting in an early series of inexpensive QSC power amps that had the bad habit of clipping asymmetrically putting DC across the loudspeakers. The popular street wisdom was that the QSC amps were too powerful for the Peavey speakers but that was not the case. QSC eventually figured it out and corrected their rogue current limiting, but since QSC was not making and selling loudspeakers back then, they were not as focussed as we were on such things.

JR

PS; Back in the 70s when I was working at VSC (variable speech control) the company making pitch shifters for sped up talking books, we did a joint project with Crown for a premium professional quality time compression/expansion box. They did a brilliant execution of our basic pitch shift invention using clever filtering to mimic the mechanical rotating head machines sample splicing. All I can say is some very smart engineering that sounded better than our SOTA.

At one stage QSC amps were nicknamed kettles - then they came good
I remember before digital sampling really started, for voice-over work I used to do manual tape edits to cut gaps between words and sentences for ads and political speech snippets at election times - a 1/4” Revox with the splicing block and scissor built into the deck was the go-to machine back then. Lot of work for 30sec or 1min of bs. Went through a lot of Chinagraph pencils and scissor blades.

 

[JohnRoberts]

The failure of IC’s in the early days has now been replaced by crumbling IC sockets from plastic degeneration for a lot of vintage gear - but at the time these pieces of gear were current the failure rate of IC’s was high so servicing was made easier. One of the problems I’ve found has been the random breakdown of contact between IC pins and the socket jaws from oxidation/tarnish or socket cracking, causing intermittent operation and sometimes HF noise.

yup. back in the 70s I would buy tl074s in 1000 pc lots and test them 100%. I ran the 4 op amps in series at unity gain inverting, but tricked them into thinking they were running at much higher noise gains with small resistors from the - input to ground. Out of 1000 piece batches I would get single digit rejects. Most rejects were not outright failures but rejected because of high noise, visible distortion, or some out of spec behavior. I used 100% testing because I was in a kit business and it helped to know that the parts tested good before my kit customers touched them.

By the 1980s I stopped testing them 100% because the production quality had improved that much.

JR

 

[RoadrunnerOZ]

yup. back in the 70s I would buy tl074s in 1000 pc lots and test them 100%. I ran the 4 op amps in series at unity gain inverting, but tricked them into thinking they were running at much higher noise gains with small resistors from the - input to ground. Out of 1000 piece batches I would get single digit rejects. Most rejects were not outright failures but rejected because of high noise, visible distortion, or some out of spec behavior. I used 100% testing because I was in a kit business and it helped to know that the parts tested good before my kit customers touched them.

By the 1980s I stopped testing them 100% because the production quality had improved that much.

JR

Yeah in those days if I needed one I’d buy 2, if I needed 10 I’d buy 15. Built up a good stock of spares! By the time the chip reliability was no longer an issue I kept up the overbuy habit and still do it today and not just with IC’s but all components - it’s proven to be a situation saver many a time.

 

[Khron]

Yeah in those days if I needed one I’d buy 2, if I needed 10 I’d buy 15. Built up a good stock of spares! By the time the chip reliability was no longer an issue I kept up the overbuy habit and still do it today and not just with IC’s but all components - it’s proven to be a situation saver many a time.

Also helps hitting price breaks

 

[RoadrunnerOZ]

Also helps hitting price breaks

Cheaper in 10’s for sure, some are 25+ for discount but I’ve found that getting the extra with the price breaks gets me a lot of free components paid for by the repair.

 

[SteveL]

I’d be checking the IC’s at the input and their rail voltages

Thanks for the suggestion...