Noise, power, voltage, case. All cause different part numbers
True, but I was more considering similar parts to the "108" range, not just "108" variants, like BC184 (TI) or ZTX102 (Zetex), though they claimed a better packaging than TO-92 as well. Or the BC204/5/6 etc. All mostly the same, and all had ABC gain variants. As for lead variants, TO-92 gave you every combination you could imagine.
And anyway, most of the variants were selected from the production line of essentially the same device apart from, perhaps, the collector epi thickness or doping in relation to breakdown voltage.
But the same occurred with power devices. Like all the TIP31 etc had a BD equivalent.
True, but I was more considering similar parts to the "108" range, not just "108" variants, like BC184 (TI) or ZTX102 (Zetex), though they claimed a better packaging than TO-92 as well. Or the BC204/5/6 etc. All mostly the same, and all had ABC gain variants. As for lead variants, TO-92 gave you every combination you could imagine.
And anyway, most of the variants were selected from the production line of essentially the same device apart from, perhaps, the collector epi thickness or doping in relation to breakdown voltage.
But the same occurred with power devices. Like all the TIP31 etc had a BD equivalent.
I think that's the likely reason though I'd say it was down to "End User" differences. The grades and testing regimes for different clients such as industrial, consumer, military, medical and aerospace instrumentation etc. often require tighter, specific degrees of reliability, parameter spread, packaging, hermetic seal etc. Usually that means different specs. and maybe pin arrays for essentially the same chip, also different packages such as TO18, TO39 rather than TO92, TO126 and many other obsolete styles too).
I've also seen five or more different 2N series transistor specifications for what is basically the same transistor so I think that the proliferation of type numbers is just as much for the end user and probably needed to distinguish different parts grades systematically, to minimise inventory/purchasing errors.
I've also seen five or more different 2N series transistor specifications for what is basically the same transistor so I think that the proliferation of type numbers is just as much for the end user and probably needed to distinguish different parts grades systematically, to minimise inventory/purchasing errors.
Last edited:
Yes, you're probably right.
I suspect the gain grades were down to industrial users complaining about the large spreads and their equipment tolerancing. Individual consumers would not have had enough clout.
Gain grades seem to have been introduced quite early on. The commonest I recall was the 2N2926 with its colour dots. Other devices were differentiated by type number, adding to a long list of thousands of devices. Many of which are now obsolete.
I suspect the gain grades were down to industrial users complaining about the large spreads and their equipment tolerancing. Individual consumers would not have had enough clout.
Gain grades seem to have been introduced quite early on. The commonest I recall was the 2N2926 with its colour dots. Other devices were differentiated by type number, adding to a long list of thousands of devices. Many of which are now obsolete.
A piece of advice required. Someone asked me where energy came from for musical transients. I said rather quickly from the DC side of a power supply and required that the amplifier would be class AB or D for transients as the person meant. For example the NAD 3020 could give quiet good transient power into two ohms.
Some maker of mains filter had stated their filter had a reserve of power for transients. For various reasons I doubted that. AC energy being not easy to store.
It made me wonder it the amplifier transformer could store energy in this sense. It must I guess.
Do class A amplifies have less good transient power. I guess they must.
Any view greatly appreciated.
Some maker of mains filter had stated their filter had a reserve of power for transients. For various reasons I doubted that. AC energy being not easy to store.
It made me wonder it the amplifier transformer could store energy in this sense. It must I guess.
Do class A amplifies have less good transient power. I guess they must.
Any view greatly appreciated.
Hi Nigel
The traditional class AB amplifiers sometimes had specifications like "peak music power". The reason behind this is that when lightly loaded, the reservoir capacitors charge to peak voltage. If they are large enough, then they can deliver power at a higher voltage for a short time. As they supply power, the voltage drops so the continuous power rating is lower.
Music often has peaks which are large compared with the average signal, so such unregulated amplifiers could benefit from the lightly loaded output voltage on the PSU.
Class A amplifiers usually draw a continuous current, and that gives rise to a lower average voltage, so in general, they are not able to deliver such a differential between peak and mean power.
Certainly some early RCA designs, and some manufacturers, exploited unregulated PSU's for Class AB amps for that reason.
Some designers recommended regulated PSU's. They give lower rail distortion and/or hum feedthrough, but the regulation stops the higher voltage peak advantage.
Desgins which transition from A to AB or B with signal might benefit, but usually the AB or B current variation can lead to several volts difference between quiescent and contniuous power output.
The traditional class AB amplifiers sometimes had specifications like "peak music power". The reason behind this is that when lightly loaded, the reservoir capacitors charge to peak voltage. If they are large enough, then they can deliver power at a higher voltage for a short time. As they supply power, the voltage drops so the continuous power rating is lower.
Music often has peaks which are large compared with the average signal, so such unregulated amplifiers could benefit from the lightly loaded output voltage on the PSU.
Class A amplifiers usually draw a continuous current, and that gives rise to a lower average voltage, so in general, they are not able to deliver such a differential between peak and mean power.
Certainly some early RCA designs, and some manufacturers, exploited unregulated PSU's for Class AB amps for that reason.
Some designers recommended regulated PSU's. They give lower rail distortion and/or hum feedthrough, but the regulation stops the higher voltage peak advantage.
Desgins which transition from A to AB or B with signal might benefit, but usually the AB or B current variation can lead to several volts difference between quiescent and contniuous power output.
Thank you John. Mostly how I saw things but the peak charging is a great way of saying it. And the general outline very helpful.
I imagine if the transformer does anything it acts as a choke apart from it's prime function. The Carver cube used a choke as a storage device they said. I guess a bit like a Cockcroft Walton multiplier of sorts.
I can't see a power filter as an add on storing energy.
Mr Old. Impedance dips noted.
I imagine if the transformer does anything it acts as a choke apart from it's prime function. The Carver cube used a choke as a storage device they said. I guess a bit like a Cockcroft Walton multiplier of sorts.
I can't see a power filter as an add on storing energy.
Mr Old. Impedance dips noted.
Hi JLH builders,
Inspired by reading posts on diyaudio I connected my old JLH amp (which I built for use as el cheapo amp in my lab) to my living room high end speakers.
I was blown away. Until now I rather used the amp to test speakers or as "spare". but this little thing rocks...
At first I was disappointed, but after 20 minutes it started to sing and after 30 minutes the music was amazing...
At the end we were listening for hours from Jazz to Classics to rock to voices...
Best regards
Christoph
Inspired by reading posts on diyaudio I connected my old JLH amp (which I built for use as el cheapo amp in my lab) to my living room high end speakers.
I was blown away. Until now I rather used the amp to test speakers or as "spare". but this little thing rocks...
At first I was disappointed, but after 20 minutes it started to sing and after 30 minutes the music was amazing...
At the end we were listening for hours from Jazz to Classics to rock to voices...
Best regards
Christoph
you're welcome, Nigel.
It might seem that a transformer should act as a choke, but really it doesn't in the scheme of a PSU. The transformer has quite a large inductance, of course, but that is due to the large number of turns on the primary and is done that way to minimise idling, or magnetising, current. When a load current is drawn from the secondary, the magnetic field reduces a little but the primary current increases to compensate, so in fact it hardly changes the overall magnetic field.
Most simple PSU's just use a bridge and a capacitor or two to smooth the ripple. These could charge to peak voltage and then drop as load is drawn, depending on the impedances (transformer winding resistances, bridge rectified voltage drop etc).
A choke is another way of reducing ripple. When used as a ripple filter the inductance is employed to oppose the ripple voltage. It is also possible to design a "swinging choke" where the core can (almost) reach saturation, at a low current,but higher currents cause the inductance to fall. However, a higher current working on a smaller inductance can still give a nearly equivalent ripple reduction.
A third way of using a choke (or another way of using the swinging choke) is to ensure it has a minimum current flow and if designed appropriately, will regulate the voltage. In this case you don't ever see the peak voltage because the choke works over the full voltage swing. so in this case you only get the average voltage output, but with a voltage drop according to the winding resistance once again.
From what I recall, JLH quite liked the idea of using a choke in his Class A (this thread) as it is the simplest means of regulating the voltage, and reducing ripple. A constant current also means a relatively simple choke design is sufficient.
In many ways I agree. Unfortunately, transformer steel seems to be another resource which is becoming depleted as it seems harder these days to get hold of. Most low current and quite a few mid-range power supplies have gone high frequency and use ferrite which has reduced the market.
Maybe those with JLH Class A's should think about choke filters.
Maybe I should add that if a choke is used as a voltage regulator, it comes immediately after the rectifier, and before the filter capacitor. If it is used after, then it (usually) just acts as an additional impedance to ripple.
It might seem that a transformer should act as a choke, but really it doesn't in the scheme of a PSU. The transformer has quite a large inductance, of course, but that is due to the large number of turns on the primary and is done that way to minimise idling, or magnetising, current. When a load current is drawn from the secondary, the magnetic field reduces a little but the primary current increases to compensate, so in fact it hardly changes the overall magnetic field.
Most simple PSU's just use a bridge and a capacitor or two to smooth the ripple. These could charge to peak voltage and then drop as load is drawn, depending on the impedances (transformer winding resistances, bridge rectified voltage drop etc).
A choke is another way of reducing ripple. When used as a ripple filter the inductance is employed to oppose the ripple voltage. It is also possible to design a "swinging choke" where the core can (almost) reach saturation, at a low current,but higher currents cause the inductance to fall. However, a higher current working on a smaller inductance can still give a nearly equivalent ripple reduction.
A third way of using a choke (or another way of using the swinging choke) is to ensure it has a minimum current flow and if designed appropriately, will regulate the voltage. In this case you don't ever see the peak voltage because the choke works over the full voltage swing. so in this case you only get the average voltage output, but with a voltage drop according to the winding resistance once again.
From what I recall, JLH quite liked the idea of using a choke in his Class A (this thread) as it is the simplest means of regulating the voltage, and reducing ripple. A constant current also means a relatively simple choke design is sufficient.
In many ways I agree. Unfortunately, transformer steel seems to be another resource which is becoming depleted as it seems harder these days to get hold of. Most low current and quite a few mid-range power supplies have gone high frequency and use ferrite which has reduced the market.
Maybe those with JLH Class A's should think about choke filters.
Maybe I should add that if a choke is used as a voltage regulator, it comes immediately after the rectifier, and before the filter capacitor. If it is used after, then it (usually) just acts as an additional impedance to ripple.
Last edited:
Thank you John. I seem to remember someone saying to me a choke fed power supply is the fashion. So as to seem wise I said it made sense.
I find writhing a question often makes me look more carefully as I become someone answering the question I asked. As I said before I am dyslexic and sometimes over value my memory, no one can remember everything. Before 2004 I wrote nothing. Seeing a machine in my mind is easier for me than many. I saw the transformer primary inductance as the key question this time.I suspect how amplifiers are thought to work and the truth may be different. Class A defines things that we would want and we should get. The power supply will always be in doubt.
Alan Blumlein would have loved a laptop I suspect. It's said he had fewer papers due to his dyslexia. At 12 he struggled to pass the 11 plus examination and needed a tutor.
I find writhing a question often makes me look more carefully as I become someone answering the question I asked. As I said before I am dyslexic and sometimes over value my memory, no one can remember everything. Before 2004 I wrote nothing. Seeing a machine in my mind is easier for me than many. I saw the transformer primary inductance as the key question this time.I suspect how amplifiers are thought to work and the truth may be different. Class A defines things that we would want and we should get. The power supply will always be in doubt.
Alan Blumlein would have loved a laptop I suspect. It's said he had fewer papers due to his dyslexia. At 12 he struggled to pass the 11 plus examination and needed a tutor.
JHL as Headphone AMP - Power consumption?
Hi everyone!
I'm new to the forum and I'm not sure if I should create a new thread for my question, but since it's JHL related here we go.
I want to build a JHL 1969 kit to use as Headphone Amplifier for my DAC, which only comes with line level output. My headphones are Sony VDR-v6 (impedance 63ohms), which still sound but perform a bit too low at line level. My question is how many watts do you think will the JHL amp consume, when amplifying around +12dB into these headphones?
Since my knowledge in electronics is quite limited (I'm a filmmaker/musician who also loves diy), I can't find the math for it. I've been planning to acquire a nice German made linear power supply I found, which comes with a toroidal transformer of 18v-0-18v 15VA. Power available would be a little bit less than that, since it would also be used to feed the linear power supply to my 5VDC DAC. But the I suppose the latter will be marginal.
Thanks a lot in advance for reading me.
Domingo
Hi everyone!
I'm new to the forum and I'm not sure if I should create a new thread for my question, but since it's JHL related here we go.
I want to build a JHL 1969 kit to use as Headphone Amplifier for my DAC, which only comes with line level output. My headphones are Sony VDR-v6 (impedance 63ohms), which still sound but perform a bit too low at line level. My question is how many watts do you think will the JHL amp consume, when amplifying around +12dB into these headphones?
Since my knowledge in electronics is quite limited (I'm a filmmaker/musician who also loves diy), I can't find the math for it. I've been planning to acquire a nice German made linear power supply I found, which comes with a toroidal transformer of 18v-0-18v 15VA. Power available would be a little bit less than that, since it would also be used to feed the linear power supply to my 5VDC DAC. But the I suppose the latter will be marginal.
Thanks a lot in advance for reading me.
Domingo
I think a 'scaled down' version would be better for headphones, running lower voltage and current. That would save you a lot of money on heat sinks and power supply. You don't need watts for headphones, rather milliwatts. I once bought a zerozone JLH based on tip31 outputs on ebay some time ago, maybe something like that would be a start. Try searching ebay or similar. I think I have seen headphone amp versions too.
Example:
https://www.ebay.com/i/122917402353...x4Ni0fDtU8r1qIRG4cvZMvzo3LSpMCpxoCUoEQAvD_BwE
Or full DIY:
The Class-A Amplifier Site - JLH Headphone Amplifiers
Example:
https://www.ebay.com/i/122917402353...x4Ni0fDtU8r1qIRG4cvZMvzo3LSpMCpxoCUoEQAvD_BwE
Or full DIY:
The Class-A Amplifier Site - JLH Headphone Amplifiers
Last edited:
Very simple designs can be used. A SE class A stage and op amp with or without loop feedback. I suspect output impedance is less important with phones. If the current booster has high gain the op amp should work in class A. A 10K resistor from op amp to -ve rail would ensure that. Try the two SE stages in antiphase of where they bottom.
Thanks a lot for your answers, definitely cleared my way. I was indeed checking out the headphones versions, but didn't know they were really any different in power consumption to the standard 2-boards kits. It is also clear to me now that 9 Watts (18V x 0,5A) would suffice for me.
@OldDIY: When you say unipolar do you mean DC? If then, would it make any difference if the current is AC? Because these boards usually work with AC power. And more importantly, how did you calculate that number, if it's easy to share.
I have seen two JHL "mini boards" using different capacitors. The ZeroZone shown by @Rallyfinnen are black. Another one around branded XD0218 are green. Would anybody have a comment on the performance of both boards? I quite like the XD since one can choose which pot to use with it, although might be manufactured with newer/cheaper components than eventually recycled ones. I wonder.
@OldDIY: When you say unipolar do you mean DC? If then, would it make any difference if the current is AC? Because these boards usually work with AC power. And more importantly, how did you calculate that number, if it's easy to share.
I have seen two JHL "mini boards" using different capacitors. The ZeroZone shown by @Rallyfinnen are black. Another one around branded XD0218 are green. Would anybody have a comment on the performance of both boards? I quite like the XD since one can choose which pot to use with it, although might be manufactured with newer/cheaper components than eventually recycled ones. I wonder.