Seeking advice on Crown DC300a speaker amp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

JW

Well-known member
Joined
Jun 8, 2005
Messages
1,100
Location
Portland USA
Hello folks,
(Schematic attached)

I have a Crown DC300a here. I've ordered replacement capacitors for the electrolytics and I'm gonna wire in a single stereo pot instead of the two mono pots.

My goal is to use this in the studio with some bigger JBL 4412 monitors.

It works fine now, but my concerns are as follows.

1. These amps have occasionally been known to blow voice coils, when a transistor fails and you end up with DC on the output.
Is there an easy way/modification to ensure that my speakers are protected? For example, I see in the service manual (too big to attach here but the first thing that comes up when you google DC300a) in one of the diagrams ,that JBL has an inline fuse between the amp and speaker (positive side), I suppose to protect from just this occurrence.  What amperage and wattage for this sort of fuse?

2. The transistors on the back on my amp don't match any in the DC300a service manual.  These are the driver and output transistors attached to the huge heatsink. In the service manual, we have one version with:

2N3773 Output
2N5804 Drivers


Apparently Crown used a few different transistors during the DC300's lifetime.

My outputs are labeled:
3990/7652

Drivers are labeled
SJ6387

Looks like several of the output transistors on one channel have been replaced with SJ6451

Anyhow,  if my goal is to use this amp in the studio with some nice JBL's, I'd like to get both channels balanced obviously, so does anyone know where I can order matched transistors of this sort, or have a recommendation in this regard? Replacements/how to match etc. ??

Thanks
 
Sides had one set up in Oceanway's "Blue Room" for years.
Those amps sound way better than the circuitry suggests.

Lets not bad-mouth quasi;
D. Self has pretty much demonstrated that it is only marginally inferior to full comp, either Sziklai or EF.

Per the DC300a, an input capacitor is smart.

If really worried about output DC, wire  2 very large electrolytics (back to back) in series with  the output (externally of course).
On paper it looks like it should sound bad, but try try it with a positive attitude.
Acoustic Research  had that exact arrangement inside their model LST,
and those speakers sounded superb (anyone here remember them circa 1971?).

 
JW said:
1. These amps have occasionally been known to blow voice coils, when a transistor fails and you end up with DC on the output.
Is there an easy way/modification to ensure that my speakers are protected?

Back when I was using a lot of these amps for PA use, each channel got a triac / SBS crowbar circuit stolen directly from the Peavey CS-800 design, as fuses are cheaper than the results of fires started by incandescent voice coils.

I built 20 of the circuits into dual banana plugs, just stack them on the amp's binding posts. Never did have one actually go DC, though.

I still have some of these amps, and thinking about using one in bridge mode for a +/- bench power supply, if I could trust a 40 year old pot not to crackle.

Gene
 
There was extensive post on diyaudio about restoring one of these amps with alternative components etc.

http://www.diyaudio.com/forums/solid-state/154256-crown-dc300a-repair-questions.html

Good luck, I also have one of these in the shed waiting for repair

Peter
 
JW said:
1. What amperage and wattage for this sort of fuse?
In-line fuses are known to be protected by the speaker and the output transistors.  :D
I've always find the compromise unsatisfactory. If you use fast fuses, they will blow too early on bass signals, make them Slo-blo and the tweeters will die first. Choosing the value is also a dilemma, because the same fuse has to protect a 100W boomer and a 10W tweeter.
Gene's suggestion is a pretty good one.  Canadian company Bruit Bleu used to have such a crowbar circuit built-in their DC300's. They were enclosed in epoxy; when we serviced their amps we had to replace all their output devices and disconnect the crowbar. For some reason thay had an LED in the circuit; one day I found one of the circuits had projected the LED into the front panel making a 1mm dent in the aluminium. A pretty good reason to put the circuit inside the amp!
Anyway, there are relay-based DC protection circuits available in kit form; I would recomment that solution. Just google "speaker DC protection circuit"

2.
Apparently Crown used a few different transistors during the DC300's lifetime.
During the time of active production of the DC300, only factory original transistors were allowed for replacement. When it went legacy, equivalence lists were released; IIRC the outputs were indeed 2N3773.

My outputs are labeled:
3990/7652 
3990 is an internal factory code; Crown used standard or semi-custom (selected) parts and had their reference put on them. Similar to what Fender do with loudspeakers.
BTW 7652 means they were manufactured in the last week of 1976.


Looks like several of the output transistors on one channel have been replaced with SJ6451
Looks also like a factory reference; 2SJ6451 is a MOSFET...


Anyhow,  if my goal is to use this amp in the studio with some nice JBL's, I'd like to get both channels balanced obviously, so does anyone know where I can order matched transistors of this sort, or have a recommendation in this regard? Replacements/how to match etc. ??
2N3773 is still current; you can get them for less than $4 a piece. Beware of replacing them with more recent/faster devices; the amp may go runaway.
 
abbey road d enfer said:
In-line fuses are known to be protected by the speaker and the output transistors.  :D
I've always find the compromise unsatisfactory. If you use fast fuses, they will blow too early on bass signals, make them Slo-blo and the tweeters will die first. Choosing the value is also a dilemma, because the same fuse has to protect a 100W boomer and a 10W tweeter.
And fuses in series with speakers will heat and cool with the signal current, changing resistance and introducing another LF distortion term.
Gene's suggestion is a pretty good one.  Canadian company Bruit Bleu used to have such a crowbar circuit built-in their DC300's. They were enclosed in epoxy; when we serviced their amps we had to replace all their output devices and disconnect the crowbar. For some reason thay had an LED in the circuit; one day I found one of the circuits had projected the LED into the front panel making a 1mm dent in the aluminium. A pretty good reason to put the circuit inside the amp!
Anyway, there are relay-based DC protection circuits available in kit form; I would recomment that solution. Just google "speaker DC protection circuit"
You want to use a robust relay. Relay contacts have been known to arc and weld closed during high current faults. You could do worse than copying Peavey for reliability/cost engineering. As I recall old school BGW amps also used a crowbar across the PS to protect amp/speakers (instead of current limiting).
During the time of active production of the DC300, only factory original transistors were allowed for replacement. When it went legacy, equivalence lists were released; IIRC the outputs were indeed 2N3773.
3990 is an internal factory code; Crown used standard or semi-custom (selected) parts and had their reference put on them. Similar to what Fender do with loudspeakers.
BTW 7652 means they were manufactured in the last week of 1976.

Looks also like a factory reference; 2SJ6451 is a MOSFET...

2N3773 is still current; you can get them for less than $4 a piece. Beware of replacing them with more recent/faster devices; the amp may go runaway.
I used 2n3773 in an amp I designed back in the early '70s.  It was robust but slow. I just found 3 data sheets (one obsolete). One claimed 5Mhz Ft and two did not even state the HF gain. I don't recall the old parts being even that fast.  My old amp design was also quasi-complementary :-[ :-[ :-[  because it was the early '70s and robust PNP power devices were not available back then.  My amp delivered something like 0.25% THD @ 20kHz which was OK for my living room (4ch @ 250W).

If your DC300 is in good condition go for it, my comment is that before investing time and treasure maybe a more modern design platform would be a good starting point. FWIW Crown pretty much replicated the old DC300 design in their later SMD based CE1000/CE2000 amps, imo a design effort driven by cost accounting, not engineering.  :'(

For perspective, your loudspeakers and room acoustics will make a far more significant impact on sound quality than the amplifier, so enjoy..

JR
 
Thanks for all the info folks.

That diyaudio thread (actually it was a link to the link that is the long resoration thread) has a lot of good info. I copied some of it below for anyone viewing this thread in the future. I should have given credit to the folks who have spoken these words, but alas, I'm too lazy to go back and do that. A google search of some of the text below should reveal that. . . .

From this thread: http://www.diyaudio.com/forums/solid-state/86650-resurrecting-crown-dc300a.html


-I am sure that there are plenty of modern transistors that will work in this amp (and I have not kept up with the state of the art) but I can say that the original amplifiers used 2N3773's. I had found a cooperative tech at Crown who told me to use ONLY Westinghouse 2N3773's and to match them by Vbe at a speciffic collector current (I don't remember ). I used Solitron millitary spec 2N3773's and they worked well.

-The reason Crown specified Westinghouse is you got epitaxial, RCA was building hometaxial. For the longest time Motorola built both, the faster epitaxial part was more money. The newer MJ15024 is from the same die family as the epitaxial 2N3773, and the data is shown on the same pages and uses the same graphs for many of the parameters. The MJ21194 is the newest version with the perforated emitter technology they got from Toshiba.

-For modern output devices look at ON Semi MJW21195 and MJW21196. Even I have a hard time blowing these up. They are not TO3, but can be made to fit on most TO3 heat sinks.
There is a similar part in TO3 packaging the MJ21193 and MJ21194. The Crown CE2000 uses the MJ21194 parts, 12 of them in each channel.
 
Back when I was using a lot of these amps for PA use, each channel got a triac / SBS crowbar circuit stolen directly from the Peavey CS-800 design, as fuses are cheaper than the results of fires started by incandescent voice coils.

Do you happen to have the schematic you could post for the exact circuit you were using?
 
lassoharp said:
Do you happen to have the schematic you could post for the exact circuit you were using?
All the guillotine-equipped DC300's I had to service had both the crowbar AND the output devices to replace, and I'm not sure if the speakers were better protected, statistically.
Speakers die of overpower more often than any other cause. I had built a batch (several 100"s) of dual-level/dual-time-constant passive relay-based protection circuits that could react fast enough for tweeters and horns. They were tuned to the speakers and built-in them. With relays, when there is an overload they release the load (via a suitable high-power resistor) so the output devices are somewhat protected.
 
abbey road d enfer said:
All the guillotine-equipped DC300's I had to service had both the crowbar AND the output devices to replace, and I'm not sure if the speakers were better protected, statistically.
Speakers die of overpower more often than any other cause. I had built a batch (several 100"s) of dual-level/dual-time-constant passive relay-based protection circuits that could react fast enough for tweeters and horns. They were tuned to the speakers and built-in them. With relays, when there is an overload they release the load (via a suitable high-power resistor) so the output devices are somewhat protected.
While this is more subtle but speakers can be more easily damaged by DC content riding on the AC signal that biases the speaker out of the center of its range of motion.

DC 300s were mostly retired by then, but Peavey had a rash of speaker failures caused by another top selling (non-Peavey) power amp that would current limit asymmetrically introducing a DC component into the AC signal. Peavey had to replace way too many loudspeakers under warranty until that amp company finally corrected their design. The customers just ASSumed (believed explanations) that the amps were too powerful for the Peavey speakers. :eek:

Capacitor coupling (or transformer) input to the DC300 could avoid inadvertent DC content coming from line level sources upstream, from damaging speakers. A fast acting crowbar at the output could be helpful, but discriminating between valid low bass and a DC fault can still leave the speaker exposed to a fault for some duration. 

JR
 
Yes, the first generation of Crown Microtech's would produce an awfully distorted signal when in protection. Due to the high HF content, HF drivers and tweeters were the sacrificial victims.
Another issue is the dynamic offset, caused by the voice-coil trying to sit where the counterelectromotive force is higher, i.e. where the induction is maximum; a number of speakers exhibited this problem, which was not well known at the time. (it was the time when 15" speakers shifted from 100W rating to 400).  As a result, portions of the voice-coil were always out of the gap and did not benefitted from radiation cooling with the pole pieces. A DC offset from the amp can result in exactly the same issue.
 
Capacitor coupling (or transformer) input to the DC300 could avoid inadvertent DC content coming from line level sources upstream, from damaging speakers. A fast acting crowbar at the output could be helpful, but discriminating between valid low bass and a DC fault can still leave the speaker exposed to a fault for some duration. 

All the guillotine-equipped DC300's I had to service had both the crowbar AND the output devices to replace, and I'm not sure if the speakers were better protected, statistically.
Speakers die of overpower more often than any other cause. I had built a batch (several 100"s) of dual-level/dual-time-constant passive relay-based protection circuits that could react fast enough for tweeters and horns. They were tuned to the speakers and built-in them. With relays, when there is an overload they release the load (via a suitable high-power resistor) so the output devices are somewhat protected.


Have either of you seen worst case scenarios for DC originating from internal failures (vs upstream sources) and do you recall the sources?

I just picked up a 300A and was hoping to use it for a set of PA monitors.
 
lassoharp said:
Have either of you seen worst case scenarios for DC originating from internal failures (vs upstream sources) and do you recall the sources?
There is a very good reason that 99.9% of audio power amps do not pass DC (music is not DC).  It is not that uncommon for line level gear to fail with DC on its output, maybe not even a hard failure.  HPF on amp inputs painlessly ignores such faults.
I just picked up a 300A and was hoping to use it for a set of PA monitors.
Good luck...  The bass response never stops, for better and worse.  8)

JR
 
Good luck...  The bass response never stops, for better and worse.  8)


I've heard a number of comments re the "good" bass response on these.    HPF it is.
 
lassoharp said:
Have either of you seen worst case scenarios for DC originating from internal failures (vs upstream sources) and do you recall the sources?

I just picked up a 300A and was hoping to use it for a set of PA monitors.
The DC300 and its sibling D150 would often die for second breakdown failure of one of the output devices. And yes, it generally results in full DC voltage at the output. IIRC the rails were +/-65V. I'm not sure, but I think the output devices were rated at 120V, unlike the new versions at 140V. Second breakdown is an issue when driving reactive loads; in a series PP amp the voltage across teh transistors is supposed to be minimum when the current is high, and vice-versa; but on a reactive load, there can be significant current when the voltage is high.
At the time the DC300 was a killer, no competition; it was a standard.
 
abbey road d enfer said:
The DC300 and its sibling D150 would often die for second breakdown failure of one of the output devices. And yes, it generally results in full DC voltage at the output. IIRC the rails were +/-65V. I'm not sure, but I think the output devices were rated at 120V, unlike the new versions at 140V. Second breakdown is an issue when driving reactive loads; in a series PP amp the voltage across teh transistors is supposed to be minimum when the current is high, and vice-versa; but on a reactive load, there can be significant current when the voltage is high.
At the time the DC300 was a killer, no competition; it was a standard.

Many amps will fail with full rail DC on the output, since the common fault is a shorted power device.  Some not all, add secondary protection to prevent damage to speakers. Fuses in rails or total amp fuse will often not be fast enough to protect the loudspeakers. Relays or fast acting crowbar clamps can help. 

This is a different issue from keeping DC out of the audio input of a DC coupled amp that would then drive DC into a speaker without having failed.

JR

PS: Yes secondary breakdown... VI limiting properly done is protective against secondary breakdown. I recall looking at the device protection in use back in the early '70s when I designed my first amp, and some were laughable, providing little if any real SOA (safe operating area) device protection, which was evidenced by some amp model's proclivity to blow up if you looked at them crosseyed (like the "Flame Linears"). 
 
JohnRoberts said:
I recall looking at the device protection in use back in the early '70s when I designed my first amp, and some were laughable, providing little if any real SOA (safe operating area) device protection
Crown were very innovative in this respect, on the first generation of Microtech's, with their  ODEP (Output Device Emulation Protection), that actually simulates the operating conditions of the output devices (using multipliers and 3 different time-constants). Unfortunately they chose to clamp the drivers, which resulted in an awfully distorted waveform, which was damageable to the tweeters and horns.


some amp model's proclivity to blow up if you looked at them crosseyed (like the "Flame Linears").
Yes; Clair Bros, we serviced their amps and speakers when they were on tour in Europe. PL700's, I think they had more spare than amps actually in use. Average was one per day...
 
Okay,

Some further research and further questions here.

For the output and driver transistors,

would there be a discernible difference between these?
MJ21194G datasheet: http://www.mouser.com/ds/2/308/MJ21193-D-108487.pdf
MJ15003 datasheet: http://www.mouser.com/ds/2/308/MJ15003-D-107529.pdf

Can someone tell me which of the output and driver transistors need to be matched with each other, or selected for some sort of minimum rating?

Also, it's been mentioned elsewhere D108 and D208 which are in the VI limiter circuit be replaced. These are 1N270 germanium diodes. Would a 1N4148 be a suitable replacement?
1N270 datasheet: http://www.redrok.com/Diode_Germanium_1N270_0.04A_DO-7.pdf
1N4148 datasheet: http://www.vishay.com/docs/81857/1n4148.pdf

Also, I think before build a relay circuit for speaker protection, I'm gonna give the two electrolytics in series a shot and see how that sounds. Any recommendations here?

And this only has to be on one side of each speaker terminal right? For example:

amp output terminal (+)  >  +leg of 4700uF (- and - tied together)  +leg of other 4700uF  >  Speaker (+) terminal
I'm assuming cap voltage rating should be at least 70+70, so 150V or more?


 

Latest posts

Back
Top