Post by Graham on May 24, 2017 8:42:07 GMT 12
I have a very basic knowledge of how a valve amp operates but thought it would be interesting if someone like Owen could explain the functions of the individual valves in a typical amp and map the flow of the signal from input to output. The valves in my Yarland Pro 88 SE are as follows with the function as described in the manual.
Rectifying Valves 2 x EL84 Signal/Voltage valve 1 x 12AX7 Impulse Valves 2 x 6SN7 Power Valves 4 x KT88
In addition of course are the transformers.
Cheers Owen, no pressure ! 
|
Post by Owen Y on May 24, 2017 10:04:18 GMT 12
Pressure! A pic of the top of the Yarland? (Or if there are details on the net - what Yarland model?) I have no formal EE training, just keep my ears & eyes open, seek good advice & inhale solder fumes often
|
Post by Graham on May 24, 2017 10:15:28 GMT 12
Ask and ye shall receive.
Yarland Pro 88 SE IV.
 
|
Post by Owen Y on May 24, 2017 12:30:06 GMT 12
Excellent thanks. From what I can see.... This Yarland appears to be a Push-Pull amp, KT88 (' Kinkless Tetrode') power valves, driven by 6SN7/6N8P triodes, with a 12AX7 triode at the front end input/amplification role. There is enough gain overall for it to be used as an 'integrated' amp - ie. able to accept 'line' level source signal levels (around say 0.3V or more, 'Input Sensitivity' is not specified). Output Transformers are designed to match 4 or 8 ohm loudspeakers. Negative Feedback is switchable on/off - which can lower distortion, improve bass control, but with a reduced power output. The interesting thing (to me anyway) with this amp is that it appears to use EL84 pentode power valves as rectifier diodes. (Conventionally an amp like this would use say a GZ34 rectifier valve - or SS rectifers.) How's that for starters?  [ Oops edit: It's unclear to me whether this amp is Push-Pull output or whether the KT88s are wired in (parallel) 'Single Ended' output mode.]
|
Post by Owen Y on May 24, 2017 12:31:07 GMT 12
|
Post by Graham on May 24, 2017 14:58:10 GMT 12
Thanks Owen.
Yes it's a push/pull amp.
So in laymans terms I gather the EL84s are converting AC to DC to be used in the circuit where needed.
The signal is first fed to the 12AX7 which could be termed a preamp. I wasn't aware what roll the 6SN7 valves played but see they control the KT88s which are the main amplifier valves. The transformers step up the signal to a suitable level for the speakers. I remember reading somewhere that amps like this are effectively running at 'full throttle' all the time, all the volume control does is alter the gain. Please feel free to correct or enlarge on my basic description.
|
Post by guitardude on May 24, 2017 15:10:53 GMT 12
Apart from having run valves amps for donkeys years, I actually know very little in the way of technical stuff about them, but i'd have thought the Yarlands would be SS rectified and that the EL84 would be used as a phase splitter maybe ? (Although again I've never seen an EL84 used in that role its usually a small signal tube like a 12AX7, so maybe the EL84 is the driver tube for the KT88 and the 6SN7 are the phase splitters ? Any chance of a CCT diagram or maybe colinf can weigh in as I think he has fixed a few of these ?
|
Post by Owen Y on May 24, 2017 17:19:59 GMT 12
So in laymans terms I gather the EL84s are converting AC to DC to be used in the circuit where needed.
The signal is first fed to the 12AX7 which could be termed a preamp. I wasn't aware what roll the 6SN7 valves played but see they control the KT88s which are the main amplifier valves. The transformers step up the signal to a suitable level for the speakers. I remember reading somewhere that amps like this are effectively running at 'full throttle' all the time, all the volume control does is alter the gain. Please feel free to correct or enlarge on my basic description.
Whoa, let's begin at the beginning Graham ..... The way I see it is like this.... Valves are amplifying devices (just like transistors). They need to be activated with a PSU voltage supply that creates an 'idling', quiescent current through each valve. When a signal voltage comes along, the valve/transistor is able to 'amplify' it - bigger voltage out. That's essentially what the 12AX7 does - because this valve has a very high 'Amplification Factor' characteristic. The 'Power output' valves (KT88 in this case) on the other hand, have the job of being capable of supplying MUCH higher voltage & current to the lspkrs - eg. KT88s can handle ~100x the current of the wee 12AX7. (Additionally, this amp operates in 'Push-Pull', which enables even more power output - but that's a separate story ). The 6SN7s are there for 3 main reasons: (i) additional amplification (ii) to 'drive' the KT88s, with more current capability than the 12AX7s can muster (iii) to 'phase-split' the signal, because this a a push-pull amp, in which a separate KT88 is used to handle each +/- 'phase' of the AC sine wave signal. (More later, on the role of Output Transformers, PSU rectification, Push-Pull, Volume Controls.... and other things I've overlooked, I will *attempt* to answer )
|
Post by Graham on May 24, 2017 20:26:26 GMT 12
OK, I'm getting the picture.
My years as an automotive engineer have made me always wanting to know how something works, which is vital for working out why it wont !
Thanks for your description and explanation. I'm sure I'm not the only one watching with interest for your next instalment.
Cheers.
|
Post by Owen Y on May 24, 2017 20:57:43 GMT 12
Ok, glad it's as clear as the proverbial so far! Rectifiers: The valves require a steady DC voltage, which needs to be generated from the AC mains supply (which is swinging +ve & -ve) by (i) 'rectifying' into +ve only voltage pulses & (ii) by 'smoothing' out those pulses into DC (Direct Current) that needs to be as steady & smooth as possible for an amplifier.. Rectification is done by using Rectifiers (either valve or SS) - which are constructed using devices called Diodes - which pass current in only one direction. This can be a tricky concept, so a picture.... 
|
Post by Owen Y on May 24, 2017 21:54:14 GMT 12
Output Transformers:Another big topic, but going back to basics again.... Unlike transistors (low voltages), valves have high voltages & high output impedance (AC resistance). As a result, valve amps do not drive loudspeaker loads (typically a low 4-8 ohms) very well. So, valve amps need an OPT to 'transform' their high output impedance (100s or 1000s of ohms) down to say 8 ohms.
|
Post by Owen Y on May 25, 2017 10:42:20 GMT 12
Assuming that not everyone has fallen asleep ..... Volume Controls:So far, we have a source input signal (say 0.5vrms) that can drive our amplifier to full output (40-50wpc in the case of this Yarland). But we do not need or want maximum output, so a Volume Control is introduced to 'attenuate' the voltage within the amp - usually near the front (preamp) end. A volume control is simply a variable resistance, a sliding/rotating/switching 'voltage-divider' (potentiometer) that divides the voltage, reducing it by dumping some of it to 'ground'. Consequently, you'll see that 'pots' have 3 connections - in/out/ground.....  [Minor aside: 'Vrms' (root mean square voltage) is technically the way to describe the 'mean' value of an alternating or swinging AC voltage - see sinewave pic above.] Please ask if any or any other topics need discussion - excuse my non-tech interpretations here & there.
|
Post by RdM on May 25, 2017 20:50:30 GMT 12
You could probably find some useful beginner (and advanced!) information here: www.tubebooks.org/technical_books_online.htm#Vacuum Tube theory & circuit design And also scroll back up to the Audio (hi-fi, amplifiers, speakers...) section. Be aware that the initial bolded title link is just an intro (see what appears when you hover) - the lower link is the full text one you want for the whole book. (The collection was started in the days of expensive bandwidth.) It seems a bit odd to me that Yarland would use a perfectly good audio output pentode, EL84, as a rectifier (pair). I just checked; their maximum cathode current rating is only 65mA. If true, maybe they had their reasons, but it still seems odd. Maybe they are driving the KT88's and it has solid state, silicon rectifiers and it's a misnomer, as guitardude wrote. Then again, it's hard to believe that such a naming mistake would occur, in the manual and on their website. I couldn't find a circuit diagram. But it definitely is a push-pull amp, despite the "SE" in its title! (Special Edition?) One way to tell would be to inspect the circuit from underneath. Identify the signal flow from valve to valve to valve that way. And the connections from the power transformer HT secondary. Perhaps a pic, with valve bases labelled? Apart from the books and the link posted above, I could just add my simplified take on operation, for fun and practise: Valves live up to their name;- they really do act like a valve. (But they're voltage devices, not really like transistors in operation which are current devices (and I don't really understand them so well;- I grew up with valves!) - except that FETs, field effect transistors, would be more like valves.) With the grid, interposed between the (grounded, say, at 0V) cathode and (positive voltage applied) anode, not connected, or is at 0V (or even positive in some rare cases), the 'valve' is turned on full bore - maximum current flow depending on power supply voltage. Just like a tap, or petrol valve say, turned full on. Nothing stopping electrons boiling off the cathode to be attracted to the anode. But if the grid is made more negative, it will (electrostatically) repel the electrons streaming off (or in a cloud around) the cathode that would otherwise be accelerated toward the anode, prevent them to some extent, and so proportionally reduce the flow... Make it far enough negative with respect to the cathode and the flow will be shut off completely. That electron flow arriving at the anode then goes through a 'load' (a resistor in a preamp stage, an OPT in a power stage. That causes a voltage drop at the anode with respect to the power supply on the other end of that load, proportional to current. It so happens then that a tiny voltage variation on the grid can cause a much larger voltage variation at the anode. Which is then tapped off, coupled through to the grid of the next stage, amplified again & etc. Ideally it's a perfectly linear relationship, input to output. In the real world, some valves are more linear than others... Which is where negative feedback can be applied. Very linear valves can get away with little or no negative feedback, and some swear by this; but in other cases it's practically needed. The opposite of "positive feedback" (like the re-amplified screech of a microphone in a PA system) - it takes a sample of the output signal and feeds it back (at the speed of light;-) to the input, reversed, to instantaneously correct the error... like magic! ;=}) There are complications and caveats with applying this, but you'd find plenty of elaboration in the textbooks linked above. In a single-ended amp, SE, the signal just goes straight through, getting larger and finally given power enough to drive speakers. The output transformer just has one primary winding (and the DC valve current flows through it, could magnetise it, less efficient) In a push-pull amp, just before the power valves, the signal is split into two mirror images - the phase splitter - each of which is applied to one of the output pair, each working across one of two mirrored halves of the primary winding of the output transformer, so they add up in the secondary. And the opposite currents in the two halves, each to the centre-tap, cancel out any magnetising effects. It's like two guys using a cross-cut saw, one pushing while the other pulls, etc. Twice the result of a single guy. Sometimes there's an additional driver or buffer after the phase splitter but before the power output valves. Finally, to use another automotive analogy, the output transformer works exactly like a gearbox. It steps down high voltage & (relatively) low current to low voltage and high current, for the speakers. Just like high rpm engine shaft with whatever torque to much lower rpm at much higher torque for the wheels. Forgive me if you knew all this... ;-) I just thought I'd see if I could explain it simply... Cheers!
|
Post by RdM on May 25, 2017 20:52:14 GMT 12
|
Post by Graham on May 25, 2017 21:16:28 GMT 12
Hi Ross
Thanks for your input and info.
I've never been able to find a circuit diagram for the Yarland either.
When I first bought it from Ben I popped the bottom cover off for a look inside and was most impressed by the neat point to point wiring and general quality of the construction and workmanship.
I will do some bedtime reading of the sites you have recommended. A lot of it will be over my head though.
By the way I still have the JVC Shibata cartridge if you are interested.
Cheers
Graham
|
Post by Owen Y on May 25, 2017 23:10:46 GMT 12
I suspect that Yarland are indeed using EL84s as rectifier valves. In the early days (1920s-30s) it was not uncommon to see the same valve specified throughout an amplifier design, including rectification. Valves were cheap & a commercial device was easier to service with just one valve type throughout. eg. Check out some early Western Electric schematics. There are designs that used 300B tubes as rectifiers. I've never seen EL84 pentodes used as a rectifier diode, but in this case it would save having to provide a 5v heater supply for a GZ34/5U4/5Z3, etc rectifier. Thus the Yarland could employ 6.3v heater supplies throughout for all valves. Graham - next time you have the bottom off, take some pics. This should reveal all.
|
Post by colinf on May 26, 2017 7:29:12 GMT 12
I agree with Owen. Take off the bottom cover and take some good photos. I could possibly help with what is going on under there. I also haven't heard of using EL84s as rectifiers but for the input stage 12ax7 and 6sn7, they could be used in that way. It would make the Yarland an unusual design.
AMR-iFi R&D
|
Post by Owen Y on May 26, 2017 9:50:23 GMT 12
Yup, as colinf sez, a decent pic or two & you'll be able draw a schematic Graham I've worked out schematics that way a few times, a good way to learn
|
Post by Graham on May 29, 2017 14:28:14 GMT 12
|
Post by Owen Y on May 29, 2017 18:18:25 GMT 12
Hi Graham - what I'm seeing is firstly: - OPTs at the front, behind the KT88s - PTX at the rear RHS nr the inputs (seems odd, but all will be revealed), opp the IEC pwoer socket. - Is that a 'bridge rectifier' block bolted to the side chassis just in front the IEC power socket? (With 4 x blue-wired spade connectors.) - Those 4 silver transformer-looking things are chokes (if they have only 2 wires) or heater transformers (or interstage transformers (unlikely)) if they have 4 wires....? - I'm seeing the KT88 heater wiring is untwisted, which says that they are likely DC heated - which is suggested also by the banks of caps in the middle - the blue caps, are they low voltage? (The brown ones are hi-voltage 450v.) - Micrprocessor or relay controlled on/off switching? - Does it have a remote-controlled motorised vol control? - Clarity Cap coupling caps. More later, better go help w dinner .....
|
Post by Owen Y on May 29, 2017 20:27:09 GMT 12
- I'm seeing the KT88 heater wiring is untwisted, which says that they are likely DC heated - which is suggested also by the banks of caps in the middle - the blue caps, are they low voltage? (The brown ones are hi-voltage 450v.) Nope - I see now that both the blue & brown large caps are hi voltage 450v.
|
Post by Graham on May 29, 2017 21:30:18 GMT 12
Yes it has a remote control for the volume.
The 2 EL84 valves are attached to the separate circuit board in the centre with the big blue and brown caps on it.
The thing with the 4 blue wired spade connectors is the on/off switch. It is in a rather awkward position down the side, but I presume they were avoiding having mains wiring passing near the valves.
No microprocessor switching, that's far too hi-tech.
Cheers.
|
Post by Owen Y on May 29, 2017 22:30:39 GMT 12
OK, so the brown bd with the 16-pin IC on it & ribbon cable to vol pot motor is the remote control receiver bd. EL84s - yup they are likley used as HT rectifiers (high tension/high voltage) & the big caps in conjunction with the chokes are the 'pi-filter' circuitry (dual mono?) - but why are the caps not identical type? Mains switch - yes correct, much better to avoid AC wiring going anywhere near the signal circuit, esp the low level input signal - you can see the input cabling travelling along the opposite side, from input sockets to front Selector switch & to the 12AX7. Are you beginning to grasp the logic of some of the layout now?
|
Post by Owen Y on May 29, 2017 22:51:27 GMT 12
PS. To me (a DIY amateur), the amp looks tidy, solid & laid out according to accepted good practice. It looks traditional all-valve, no fancy SS circuitry for biasing or PSU regulation, valve loading or such. I see a couple of chunky diodes on the rear edge of the central PCB - they may the the rectifiers for the valve heaters 6.3vdc supply. And the 2 'axial leaded caps lying in the middle of the board are the associated DC smoothing (filter) caps. So, Sherlock, we are gathering more clues
|
Post by Graham on May 29, 2017 22:56:23 GMT 12
Thanks Owen. I'm starting to get my head around how it all works.
Do you think maybe it is a Chinese clone of some early design or a one-off ?
I continue to be impressed by the sound it produces.
|
Pundit
Post by Gryffles on May 29, 2017 23:02:21 GMT 12
Good thread guys.
What are the cylindrical shaped things located near each KT88 socket? (black on the outside with white internals)
|
Post by Owen Y on May 29, 2017 23:05:42 GMT 12
Most modern valve amps follow traditional tried-and-true circuitry - so yes, 'clones', but not necessarily a bad thing. Some modern amps will incorporate modern tricks, in order to 'improve' the amp technically (say distortion) or for convenience (eg. auto biasing) - but additional complication can lead to unreliability. I'd like to know what's in the 4th box in front of the IEC power socket, behind the OPT boxes.
|
Pundit
Post by Gryffles on May 29, 2017 23:08:37 GMT 12
Thanks Owen. I'm starting to get my head around how it all works. Do you think maybe it is a Chinese clone of some early design or a one-off ? I continue to be impressed by the sound it produces. I thoroughly enjoyed it too Graham. Sounded very, very good with a quad of Shuguang Black Treasure KT88 that I borrowed. Not a subtle difference from memory.
|
Post by Graham on May 29, 2017 23:13:32 GMT 12
Yes the same thought occurred to me. There are only 3 big transformers. Maybe it is just for a balanced look and has nothing significant under it. I considered lifting the cover/box off but accessing the screws would involve too much dismantling of circuit boards etc, so I left it alone. We will never know !!
|
Post by Owen Y on May 29, 2017 23:27:50 GMT 12
I'd like to know what's in the 4th box in front of the IEC power socket, behind the OPT boxes. It's possibly a separate PTX for the valve heaters supply - you can see the thick black wires emerging underneath, heading for the central PSU board. Note that these wires are much thicker than the blue & white wires coming out of the main PTX alongside. That's because the main PTX supplies the KT88s which each draw around 100mA static/quiescent current when 'idling' - times 4, that's only 400mA (0.4A at maybe 500vdc). The 6SN7s & 12AX7 draw maybe only another 25mA total. Whereas the heaters for all 7 signal valves total around 8A (at 6.3vdc) - so, fat current-carrying wire required from the heater PTX.
|
