Marko25,
Simplicity. Keep it simple for you first project. You can get sophisticated later - although you might not want to.
3 signal-path transistor, single-ended class A. 1 or 2 amps bias current. Big heatsinks. No adjustments.
It will be an ear-opening experience compared with conventional push-pull.
Not very loud, though.
Brian
Simplicity. Keep it simple for you first project. You can get sophisticated later - although you might not want to.
3 signal-path transistor, single-ended class A. 1 or 2 amps bias current. Big heatsinks. No adjustments.
It will be an ear-opening experience compared with conventional push-pull.
Not very loud, though.
Brian
Thanks Brian, just what I am after achieving to be honest, maximum detail. As I mentioned in an earlier post, I don't really listen too loud. It may just be that I do a scaled down version of what I was originally looking at as a tester, you can't get much simpler than a single ended class A from the schematics I've seen. I only proposed push pull as it split the transistor and hence the loading/heat dissipation from a single transistor however if I go scaled down to start this will be less of a concern and cost.
You have probably seen circuits like:
http://diyaudioprojects.com/Solid/IRF610-Class-A-Headphone-Amp/
and the variety of amp/ kits and modules available commercially.
My 1st Class A amp was a SWTPC (70's) headphone amp.
I resorted to running a fan on it.
There is a general understanding in Semiconductors:
For every 10 degree (C) increase in operating temperature above ambient temp, the life of a device decreases by 50%.
http://diyaudioprojects.com/Solid/IRF610-Class-A-Headphone-Amp/
and the variety of amp/ kits and modules available commercially.
My 1st Class A amp was a SWTPC (70's) headphone amp.
I resorted to running a fan on it.
There is a general understanding in Semiconductors:
For every 10 degree (C) increase in operating temperature above ambient temp, the life of a device decreases by 50%.
Marko,
The tube amp suggestion is actually a good one.
The "natural" harmonics are not an issue with a design that is not intended to create "extra" harmonics. Some tube designs use methods that imho create a euphony that is nice, other tube designs are more aimed at "accuracy" - but having said that, it is not always clear which is which. Also, pentodes + feeedback imho do not lead to something that I like to listen to.
The high voltage is not a big deal, just don't stick ur hand in there when it is on! Otherwise it is actually an easier build and adjust than a solid state amp. (fewer things that let the "magic smoke out".)
For a solid state build you really need a scope and a variac with ammeter at minimum. You can use a PC or CD player as a signal generator (burn a CDR with what you need... eh?)
What you want in terms of "transparency" may not come from the place(s) you expect. There are some clear benefits to class A in both tubes and solid state, plus some issues (heat and power being the most obvious).
No matter what amp you build, if you want to optimize the results without getting too exotic, one of the easiest and most important things you can do is to run separate supplies for the "board" and for the outputs. You can just run a regulator for the "board" but there are some questions about the "sound" or lack thereof of any "regulator" - you can get into the series vs. shunt for starters... but no matter what you'll get better performance by having separate supplies. The simplest method is to use a "cap multiplier" for the "board" supply... This technique will improve the "transparancy" on chreshendos. Big bang for little bux.
Within reason, maximize the power supply, say up to 2x (or more)the ordinary current (amperage) spec is not a bad idea at all.
Dual Mono is not a bad idea either.
You also want to consider the differences between BJTs (bipolar transistors) and Mosfets. Also the differences between BJTs in the input stage and JFETs.
In my experience it makes a difference, although it is difficult to "pin it down" to specific sounds or via measurements.
Its a big can of worms you've stepped in to.
So, the thing to so is to just pick a relatively simple and well known design that you can get or make some PCBs for without going nuts and build up the first decent Class A amp you can get your hands on.
Then build another one based upon what you have learned about the circuit and what you hear... do the power supply bit I suggested and you'll maximize the result whatever it is, and have fun! Don't spend too much time on the first one worrying about the details and "gilding" it. Make it work, test it, listen to it. Go on to the next.
That's the plan!
_-_-bear
The tube amp suggestion is actually a good one.
The "natural" harmonics are not an issue with a design that is not intended to create "extra" harmonics. Some tube designs use methods that imho create a euphony that is nice, other tube designs are more aimed at "accuracy" - but having said that, it is not always clear which is which. Also, pentodes + feeedback imho do not lead to something that I like to listen to.
The high voltage is not a big deal, just don't stick ur hand in there when it is on! Otherwise it is actually an easier build and adjust than a solid state amp. (fewer things that let the "magic smoke out".)
For a solid state build you really need a scope and a variac with ammeter at minimum. You can use a PC or CD player as a signal generator (burn a CDR with what you need... eh?)
What you want in terms of "transparency" may not come from the place(s) you expect. There are some clear benefits to class A in both tubes and solid state, plus some issues (heat and power being the most obvious).
No matter what amp you build, if you want to optimize the results without getting too exotic, one of the easiest and most important things you can do is to run separate supplies for the "board" and for the outputs. You can just run a regulator for the "board" but there are some questions about the "sound" or lack thereof of any "regulator" - you can get into the series vs. shunt for starters... but no matter what you'll get better performance by having separate supplies. The simplest method is to use a "cap multiplier" for the "board" supply... This technique will improve the "transparancy" on chreshendos. Big bang for little bux.
Within reason, maximize the power supply, say up to 2x (or more)the ordinary current (amperage) spec is not a bad idea at all.
Dual Mono is not a bad idea either.
You also want to consider the differences between BJTs (bipolar transistors) and Mosfets. Also the differences between BJTs in the input stage and JFETs.
In my experience it makes a difference, although it is difficult to "pin it down" to specific sounds or via measurements.
Its a big can of worms you've stepped in to.
So, the thing to so is to just pick a relatively simple and well known design that you can get or make some PCBs for without going nuts and build up the first decent Class A amp you can get your hands on.
Then build another one based upon what you have learned about the circuit and what you hear... do the power supply bit I suggested and you'll maximize the result whatever it is, and have fun! Don't spend too much time on the first one worrying about the details and "gilding" it. Make it work, test it, listen to it. Go on to the next.
That's the plan!
_-_-bear
Well, I've listened to quite a few speakers through my amp and by far the speakers I have in my opinion let the maximum amount of detail through to suit my taste. If it is my speakers I'd have to spend a hell of a lot more or get into horns etc to potentially get to were I'm going so I'd say at this point its the amp, its not really very good but adequate.
Cambridge Audio 340A
I'm not saying 'don't build an amplifier'.
I've built many amplifiers. I'll build an amplifier at the drop of a hat. I have parts here for amplifiers in case I feel the need for building one coming on.
You can hear ghosting thru your 'phones that you can't hear thru your speakers. You don't say, but you must be using an amplifier with the phones, probably you plug them into your integrated amp.
Now maybe the headphone section of the 340A is head-and-shoulders above the rest, but probably not...
No, the chances are that the differences you hear are down to the massive advantage your headphones have over your speakers. Don't forget that the phones improve the SNR you experience by cutting out extraneous noise (it's quieter with them on).
'Course I could be wrong, you could have a separate rig with an expensive headphone amplifier and even a different source, but then you should state that if you're making comparisons.
w
Phones really are in a different league from speakers in terms of fidelity, they just have some inconvenient down sides.
I've built many amplifiers. I'll build an amplifier at the drop of a hat. I have parts here for amplifiers in case I feel the need for building one coming on.
You can hear ghosting thru your 'phones that you can't hear thru your speakers. You don't say, but you must be using an amplifier with the phones, probably you plug them into your integrated amp.
Now maybe the headphone section of the 340A is head-and-shoulders above the rest, but probably not...
No, the chances are that the differences you hear are down to the massive advantage your headphones have over your speakers. Don't forget that the phones improve the SNR you experience by cutting out extraneous noise (it's quieter with them on).
'Course I could be wrong, you could have a separate rig with an expensive headphone amplifier and even a different source, but then you should state that if you're making comparisons.
w
Phones really are in a different league from speakers in terms of fidelity, they just have some inconvenient down sides.
Marko25 said:
Sounds like solid advice Bear, thanks a lot, I'm becoming so tempted by the tube idea, I've posted in that forum as well.
Edit: someone just suggested I would need 60WPC tube to feed my speakers so they suggested a tubed pre-amp
You may be surprised how loud, say, a 5W amplifier sounds. I am not an experienced tube designer, but I think the old fashioned tube is one of the best amplifying devices you can use in terms of musical integrity. Unfortunately, they usually use output transformers to make up for the absence of p-channel tubes. However, it might be that if you make a single-ended tube output you can dispense with the transformer. Shocking voltages involved, though.
I know, I'm too stubborn to be put off that easily
Yes I am, I think I mentioned it in a previous post.
Its only me, the wife and the cat, none of which make much noise, I only wear the headphones when she's watching TV or something so I probably have the most background when listening to them .... The main difference with the headphones of course is the sensitivity compared to my speakers which could be the issue but I'll still be building some fine amps before I think about changing them.
I would point out that the ghosting is not faint or distant, its quite obvious so you can't really miss it.
Thanks for all the replies
traderbam said:
I didn't mention that tubes are usually high V low I devices, so the other reason a transformer is used is to match the relatively low V high I requirement of speakers. I have no idea if there exists a lowish V tube that can swing a few amps or not. I suppose you could parallel them up, but this gets complicated and expensive.
Marko25 said:
So I went looking for a review of this speaker. Not being an audiophile, I don't read stereophile.
$5450 is what I discovered. Is that right? Ballpark? This speaker costs thousands of dollars? $4000 discounted?
88dB is what I also discovered.
Together these might explain a few things.
1. It would explain why you would think these speakers ought to be good enough to reveal detail you can hear with headphones.
2. It might explain why you can't.
88dB is an inefficient speaker by any standards. Your Cambridge Audio amplifier is an excellent budget HiFi amplifier but at 40 Watts it is grossly undermatched to these speakers, i.e. TOO QUIET. Why would you be operating these speakers with an amp that cost less than 10% what they did?
These speakers are really for operation with higher-power solid-state amplifiers. They come from the 'brute force' school of engineering. If you put enough power into them to hear the ghosting, however, you may upset your neighbours with the peaks. Headphones really are different from speakers, in this respect too.
Forget valves, they're on such a distant branch of the evolutionary tree they can no longer breed with these speakers.
w
Re: Re: Class A amplifier build (need a lesson in audio engineering)
Well £2250 discounted for me, there made in sheffield which is just down the road from me.
I purely have the amp from my old system pre speaker upgrade, I couldn't afford the whole hog in one go. personally I don't find wanting for volume and rarely have it above halfway on the volume setting.
Do you really think its just a power thing?
Thats really the reason I was proposing matching the RMS i.e. 100 watts at the start of this thread. I partially wondered about the power.
Saying that though someone is using similar sensitivity speakers with a 8 watt tube on my tube thread with great success apparently.
Well £2250 discounted for me, there made in sheffield which is just down the road from me.
I purely have the amp from my old system pre speaker upgrade, I couldn't afford the whole hog in one go. personally I don't find wanting for volume and rarely have it above halfway on the volume setting.
Do you really think its just a power thing?
Thats really the reason I was proposing matching the RMS i.e. 100 watts at the start of this thread. I partially wondered about the power.
Saying that though someone is using similar sensitivity speakers with a 8 watt tube on my tube thread with great success apparently.
plate tectonics
I think there is still some life left in tube technology.
The thing is, Marko, you seem to be looking for an easy to build, near-audiophile quality amp. I think this is achievable using either tubes or SE transistors. But it is really hard to make a similar quality PP transistor amp that is easy to build. Experiment by all means, but I think it would be handy to make some low power, good reference amps before you go on to do battle with PP silicon.
I think there is still some life left in tube technology.
The thing is, Marko, you seem to be looking for an easy to build, near-audiophile quality amp. I think this is achievable using either tubes or SE transistors. But it is really hard to make a similar quality PP transistor amp that is easy to build. Experiment by all means, but I think it would be handy to make some low power, good reference amps before you go on to do battle with PP silicon.
A great deal depends on the size of your room and what distance from the speakers you listen at. Also, the type of music you will be listening to...
I'm going to say that while in theory headphones may be more "accurate" than you speakers, they may not be so in some cases.
I'd say that using the Cambridge you have no idea what either the speakers nor the headphones are capable of.
Next idea, there's nil difference in difficulty between building a first class headphone amp (not a bad idea at all) and a power amp. Just that the power amp is bigger and heavier. About the same complexity (assuming we're not talking chipamps here, right?).
Making a tube amp that will actually put out enough juice to run your speakers properly will be a relatively big size project as to get good fidelity you will sink some British Pounds Sterling into the iron at minimum. To do a credible job, imho you'll need triodes, and that means big tubes, not little ones. That means 211, 845, or 811. The latter are affordable at least. The chinese make decent versions of the first two. You can do it. But you need good output iron. Sowter makes nice iron, so does Lundahl. Expensive in that power.
Solid state you can use 1/2 the iron (no output iron) and get pretty good results if you pick the right circuit.
Keep in mind, that it's GIGO all the way. The better you make things, it seems, the more little defects start to become annoying.
So you won't get glorious sound with a very excellent power amp and a so-so preamp and so-so CD source.
But, skip the decision worries and just PICK SOMETHING and build it. Start with something that won't wreck your budget, but has a chance of sounding good based on specs, design, and reports from others (which may or may not be meaningful at all).
Consider biamping, or triamping if the speaker supports it.
A good excuse for building more than one amp.
Also, if the speakers are made nearby, go to the salons that carry them, and maybe to the factory and see what they use and if it sounds worth a darn... if so take note of the gear and make a determination as to what the topology of the circuit is!! You can come here and post it, and probably you can get a run down on the innards of almost anything on this forum. Then pick something that does more or less what that does, and build it!
That's a few ideas there...
_-_-bear
I'm going to say that while in theory headphones may be more "accurate" than you speakers, they may not be so in some cases.
I'd say that using the Cambridge you have no idea what either the speakers nor the headphones are capable of.
Next idea, there's nil difference in difficulty between building a first class headphone amp (not a bad idea at all) and a power amp. Just that the power amp is bigger and heavier. About the same complexity (assuming we're not talking chipamps here, right?).
Making a tube amp that will actually put out enough juice to run your speakers properly will be a relatively big size project as to get good fidelity you will sink some British Pounds Sterling into the iron at minimum. To do a credible job, imho you'll need triodes, and that means big tubes, not little ones. That means 211, 845, or 811. The latter are affordable at least. The chinese make decent versions of the first two. You can do it. But you need good output iron. Sowter makes nice iron, so does Lundahl. Expensive in that power.
Solid state you can use 1/2 the iron (no output iron) and get pretty good results if you pick the right circuit.
Keep in mind, that it's GIGO all the way. The better you make things, it seems, the more little defects start to become annoying.
So you won't get glorious sound with a very excellent power amp and a so-so preamp and so-so CD source.
But, skip the decision worries and just PICK SOMETHING and build it. Start with something that won't wreck your budget, but has a chance of sounding good based on specs, design, and reports from others (which may or may not be meaningful at all).
Consider biamping, or triamping if the speaker supports it.
A good excuse for building more than one amp.
Also, if the speakers are made nearby, go to the salons that carry them, and maybe to the factory and see what they use and if it sounds worth a darn... if so take note of the gear and make a determination as to what the topology of the circuit is!! You can come here and post it, and probably you can get a run down on the innards of almost anything on this forum. Then pick something that does more or less what that does, and build it!
That's a few ideas there...
_-_-bear
My rooms about 4.5 x 4 metres, no over sized by any means. Mainly I listen to rock music but also listen to classical, vocal and acoustic hence the speaker as this copes brilliantly with all of these.
When I got the speakers I listened through an AYRE amp which sounded great, its just that I was unable to listen to the particular records I refer to as at that point I hadn't spotted it so I don't know if they allow this sound to come through.
No were not talking chipamps, As I need a headphone amp anyway I think I may just bite the bullet and build one of those first, then move onto an integrated or pre amp & power amp separate.
Completely agree, the only thing after my amps will be the RIAA Phono stage to build...but thats a whole other thread
Okay, okay I'll just take the plunge on a head amp first
Thanks for your help.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.