Okay so I have this industrial amplifier that uses this circuit: http://www.kn34pc.com/sch/sch_ampl/ems_1000/sch_ampl_resprom_emc_1511_cr.jpg
I am using it with B&W DM310 speakers and sources are laptop, tuner and turntable.
I like the overall sound of this amplifier. Fast and punchy bass with very good lows, even with the 8 inch woofers. Generally a very detailed sound also.
More info on this amplifier can be found here: KN34PC-Silistra
Now this is supposedly a class B amplifier with a class A pre-amplifier. Now this of course bring in the issue of crossover distortion. And this amplifier has it. It's not too bad at lower frequencies, but mid-highs sound very fatiguing.
For an example on the song "Wishing Well" by Airborne Toxic Event, the beginning has this... synth?? tune and it sounds very harsh. That is just one example, but anything like that sounds bad with this amplifier.
So what I want to know is if there is any way to improve this circuit to reduce crossover distortion? Because I really like the overall sound of this amplifier and I don't have enough money currently to buy another amplifier.
I've already re-capped it and put in new wiring. This did help with overall SQ but not with distortion.
Also, the input of this circuit is a bit odd. First you have a 3.9K resistor, a 470p input to ground bypass cap, then a 470n coupling cap, the feedback resistor, and then another coupling cap. Is this good or should it be changed?
Thanks! 🙂
I am using it with B&W DM310 speakers and sources are laptop, tuner and turntable.
I like the overall sound of this amplifier. Fast and punchy bass with very good lows, even with the 8 inch woofers. Generally a very detailed sound also.
More info on this amplifier can be found here: KN34PC-Silistra
Now this is supposedly a class B amplifier with a class A pre-amplifier. Now this of course bring in the issue of crossover distortion. And this amplifier has it. It's not too bad at lower frequencies, but mid-highs sound very fatiguing.
For an example on the song "Wishing Well" by Airborne Toxic Event, the beginning has this... synth?? tune and it sounds very harsh. That is just one example, but anything like that sounds bad with this amplifier.
So what I want to know is if there is any way to improve this circuit to reduce crossover distortion? Because I really like the overall sound of this amplifier and I don't have enough money currently to buy another amplifier.
I've already re-capped it and put in new wiring. This did help with overall SQ but not with distortion.
Also, the input of this circuit is a bit odd. First you have a 3.9K resistor, a 470p input to ground bypass cap, then a 470n coupling cap, the feedback resistor, and then another coupling cap. Is this good or should it be changed?
Thanks! 🙂
This amplifier may not be suitable for other than a subwoofer or LF biamping, since the heat sinks are likely not adequate for higher bias current.
I wouldn't change the input circuit, other than perhaps installing better small capacitors.
Last edited:
We could try higher bias, since the heatsinks are pretty huge and as it is now, this amplifier emits next to nothing in terms of heat, even at high output for long periods of time (yes, I like it loud 😛 )
How do I set higher bias? I think I have some sort of a clue but I'd rather hear from someone smarter than me, so please, share with me your wisdom 😀
I replaced input electrolytics with film. Same size as original, 470n and then 1uF. Should I go higher maybe?
Thanks! 🙂
It appears that the bias in not adjustable in this unit, not surprising in a class B amplifier.
Yes, I noticed.
Let me point out that I have basic knowledge about electronics and how amplifiers work. So I basically know what most of the components do. But I'm still learning.
So how should I change the bias circuit?
V168 looks to be another part of the bias circuit.
Also are there any other changes that could be done to this circuit to get better sound? Other than bias.
That schematic is an eye bulger. The bias circuit would take a lot of analysis and reiteration to change just right, and would not get done without the output section running away at least once, which could easily leave you better off starting from scratch. That is a class G amp with a Sziklai pair on one side for all NPN quazi-complimentary power outputs. The sinks have been sized to handle the possible dissipation expected with the bias the way it is. The whole design (including layout) is contrary to requirements for stellar HF performance. The same circuit with modern parts could probably perform much better but would require that every value be changed. Anyway, if you're that fresh and don't care whether you lose what that museum piece is currently capable of, here's a nice article I recently saw on the kinds of things you have to think about for stable class AB bias: http://hifisonix.com/wordpress/wp-c...on-for-Audio-Amplifier-EF-Triples-V1.0231.pdf
Last edited:
Almost certainly reducing the impedance of the supply. More caps, close-in to the output devices in other words.
Thank you for the link, Andrew. It's a bit of a longer read so I'll read it a bit later, busy with the holidays right now 🙂
Anyways, I think maybe I should take the amplifier modules out and measure all the resistors, etc... in the bias circuit to make sure everything in within tolerance and working to it's full potential? 🙂
Also, should I bump the input coupling cap size? I have some nice 4.7uF K73-16V caps that I could throw in there.
This amplifier's power supply has separate caps for both modules. 27V lines have 2000uF (1000x2) per rail and 48V rails have 4700uF per rail. I think all the caps are fine because this thing has some serious bass performance and there is next to no output hum. I could be wrong tho, so I guess I'll replace the 1000uF caps. The bigger ones should be okay. They are rated for 72V, so there is a lot of headroom there.
Anyways, I think maybe I should take the amplifier modules out and measure all the resistors, etc... in the bias circuit to make sure everything in within tolerance and working to it's full potential? 🙂
Also, should I bump the input coupling cap size? I have some nice 4.7uF K73-16V caps that I could throw in there.
This amplifier's power supply has separate caps for both modules. 27V lines have 2000uF (1000x2) per rail and 48V rails have 4700uF per rail. I think all the caps are fine because this thing has some serious bass performance and there is next to no output hum. I could be wrong tho, so I guess I'll replace the 1000uF caps. The bigger ones should be okay. They are rated for 72V, so there is a lot of headroom there.
Last edited:
40 year old edge connectors, switch and relay contacts, hand laced wiring. That unit is a serious project even hoping for original performance, which is very clearly specified as fairly narrow band and otherwise limited for modern hi-fi applications. It's probably simply not capable of doing much more than it was originally designed for without Complete overhaul. Just looking at some of the component values suggests that it was thoroughly tweaked to do what it could as it came out of the shipping box, but after so much time and who knows what duty before you got hold of it, all bets are off. All the semiconductors are probably obsolete and maybe even no chance of getting datasheets for them. Seriously, I'd keep a soldering iron as far away from it as possible..
Last edited:
So I would simply be better off saving money for a new amplifier/amplifier build?
I understand your criticism against this amplifier but it's not that bad! Honestly, if it wasn't for the crossover distortion, this amplifier would sound excellent! You said that it should have bad HF performance, right? Well, the distortion aside, the highs are there. I wouldn't call it a bad sounding amplifier, it's actually much better than the one I had before. I used to have a Samsung L4A. Absolutely terrible. All tone controls flat and no loudness, this thing sounded terrible. Muddy, no bass, no highs, just awful. This one however, sounds alright.
So before it gets gutted and hits the dump, is there ANYTHING I can do to get it to sound better before I get enough cash for a new amplifier?
I understand that I could beef up the power supply, but is there anything cheaper?
If I were to build a new amplifier, how would you comment on this: http://www.ebay.com/itm/audio-power...859?pt=LH_DefaultDomain_0&hash=item3a8dd3008b
I could reuse the case and transformer 🙂
Last edited:
"So before it gets gutted and hits the dump"
That's pretty much the deal, as it's happened to me many times before with used equipment that was a lot closer to cutting edge than that thing is now. Truth is technology has advanced to the point where it's quite economical to leave that thing in the dust performance-wise. Just using the old chassis would be a painful exercise, as it's laid out for functions you don't even need. and the sinks are punched or drilled for TO-3s. The power transformer is probably the only part handy for a new design. Even a 3886 chipamp is clearly better, though I would suggest you buy a kit from one of the designers/suppliers out of this forum rather than an Ebay job. It reads like you're up to it, so maybe you would want to investigate some of the great discrete designs taken to purchasable PCBs and BOMs from here as well. It would simply take less work than refurbishing your salvaged boat anchor, and sound a Lot better in the end.
That's pretty much the deal, as it's happened to me many times before with used equipment that was a lot closer to cutting edge than that thing is now. Truth is technology has advanced to the point where it's quite economical to leave that thing in the dust performance-wise. Just using the old chassis would be a painful exercise, as it's laid out for functions you don't even need. and the sinks are punched or drilled for TO-3s. The power transformer is probably the only part handy for a new design. Even a 3886 chipamp is clearly better, though I would suggest you buy a kit from one of the designers/suppliers out of this forum rather than an Ebay job. It reads like you're up to it, so maybe you would want to investigate some of the great discrete designs taken to purchasable PCBs and BOMs from here as well. It would simply take less work than refurbishing your salvaged boat anchor, and sound a Lot better in the end.
I just thought of something. If you're interested in getting the most value out of that old amp, leave it exactly as it is so you can listen to it compared to the new one you build!
Hi,
No, if the indicated 0V across the base resisistor is correct.
AS AE says the schematic is an eye bulger, and it makes
my head hurt trying to follow it. However clearly V1 to
V4 are mounted on an external heat sink, and so is K1.1,
whatever that is, though I'm pretty lost as to the point
of the V142 and V147 optocoupling, with no isolation.
Something simpler would be better, and I totally disagree
you can't make a fine sounding amplifier with 4 TO3's
per channel, you can make a fabulous sounding amplifier.
In fact given the OTT nature of the circuit you could probably
build a simpler circuit, with optimum biasing and optimum
compensation with using mostly the original transistors.
rgds, sreten.
Last edited:
Thank you, Andrew, for the good advice!
My problem with building a new amplifier is that I don't have a lot of money, and that's why I wanted to go with an Ebay kit.
The chassis of this amplifier is fairly easy to reuse, it comes apart very easily and is easy to customize. What I want to to is use the frame as a base and then build a wooden chassis around it.
Or I could build an entirely new chassis and just use this amplifier as a part of my PA setup for bass horns.
Also, how much is the average price of a decent LM3886 kit? I need one cheap 😀
Dedicated designs are pretty much out of my budget for now.
My problem with building a new amplifier is that I don't have a lot of money, and that's why I wanted to go with an Ebay kit.
The chassis of this amplifier is fairly easy to reuse, it comes apart very easily and is easy to customize. What I want to to is use the frame as a base and then build a wooden chassis around it.
Or I could build an entirely new chassis and just use this amplifier as a part of my PA setup for bass horns.
Also, how much is the average price of a decent LM3886 kit? I need one cheap 😀
Dedicated designs are pretty much out of my budget for now.
Using the original small signal transistor would be a pain in the butt, since this thing is built in Bulgaria and doesn't use the "standard" western parts.
But the outputs could be used, I guess.
V1, 2, 3, 4 are all mounted on a heat sink. They are Tesla KD502 nad KD503. Both channels have two of each.
Last edited:
I didn't mean that the TO-3 outputs were part of why this amp could easily be beat, I just think re-using the old ones could be worth no more than the education, assuming you could find out anything else more descriptive about them. If you wanted to install another circuit that uses 4 TO-3s per side that would be okay, but possibly just another limiting/restricting factor, depending on demands and time/cost budget. You could run a new design class G circuit with higher bias. Sure, that'd be everything but easy. There probably isn't enough sink for a high bias class AB with the original supply unless the user were to guarantee 8 Ohm loads and possibly judicious volume settings. If you wanted to put some cheap LM3886 boards in there, you optimally have to mount a blank spreader plate over the original sink holes, for example. +/-50 volts is too high for that chip anyway. Maybe you could squeak a TDA7293 in there. I still think saving it to compare against later might be the most to be had out of this amp eventually, aside from it meanwhile being useful as it is . After listening comparison you might choose to salvage some parts of it for another build. The problems with the Chinese chip amp boards are well documented on this forum as finding a way to somehow end up costing more than the US sourced designs and parts.
Last edited:
I see. Well, since class G seems not to have very good SQ, I'd rather not bother with it but build a new class AB circuit.
Now my concern with chipamps is the sound quality. I need something that will sound good with my B&W DM310 speakers. I know the LM3886 and TDA7293 are good, but are they good enough?
So as it is now, there is nothing that can be done with this amplifier other than using it as a doorstop or a paperweight?
" I know the LM3886 and TDA7293 are good, but are they good enough?"
I think there's a good chance, if this amp seems to have useful characteristics to you as is, that a TDA7293 would be good enough. There are parallel mono 7293 circuits available on Ebay that could potentially load the supplies to the limits with a 4 ohm load, but you'd likely run into problems with sink overheating if you did that, while 8 ohms might be okay. There are the problems of off the cuff PCB design and counterfeit chips from Ebayers, but you could take the risk if you wanted, or purchase from more reputable suppliers. Anyway, as I said, you can't know how another amplifier sounds compared to the one you have now if you destroy or rework the existing one first. I think it would be most worthwhile to have the comparison experience, whatever you chose to do for new amps. Class G can have very good SQ, especially if for most of your listening you never actually break into the second rail, and only use it for dynamic headroom, although 48 volts for a high rail is not very ambitious for a power amp these days, unless low impedance speakers are used.
I think there's a good chance, if this amp seems to have useful characteristics to you as is, that a TDA7293 would be good enough. There are parallel mono 7293 circuits available on Ebay that could potentially load the supplies to the limits with a 4 ohm load, but you'd likely run into problems with sink overheating if you did that, while 8 ohms might be okay. There are the problems of off the cuff PCB design and counterfeit chips from Ebayers, but you could take the risk if you wanted, or purchase from more reputable suppliers. Anyway, as I said, you can't know how another amplifier sounds compared to the one you have now if you destroy or rework the existing one first. I think it would be most worthwhile to have the comparison experience, whatever you chose to do for new amps. Class G can have very good SQ, especially if for most of your listening you never actually break into the second rail, and only use it for dynamic headroom, although 48 volts for a high rail is not very ambitious for a power amp these days, unless low impedance speakers are used.
Last edited:
- Status
- Not open for further replies.