• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Some help with PCL86 Headphone amp Project

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Now the only question is, what are the pros and cons of OTL vs OT?
All the performance/quality pros go to the OTL, the only advantage of the OT being its impedance adaptation capability, which is essential with low to medium power tubes having peak output currents in the 10's ~ 100's mA (at most) driving a low impedance load such as a speaker.
With headphones, the situation is different: even 20mA rms into 25Ω (which is about the lowest you can find) results in 500mW, literally a deafening level for headphones. In these conditions, an OT is completely dispensable.
 
What is the difference in sound OTL vs OT?
I am not going to venture on this ground: I think a number of people will, and they will probably speak in favor of the OT, which is an essential component (not to say the only significant one) of the "tube sound".
Directly or indirectly, a transformer generates more distortion than all of the other sources, and therefore colors the sound in an unmistakable way.
In an OTL, tubes are simply ordinary active components, with some peculiarities not unlike latFETs for instance, but not quite. The final characteristics can match those of any semi amp: bandwidth of 1Hz to 1MHz, THD an arbitrary 0.0XXX1% if that is your choice....
As soon as you put a transformer in the loop, things become different, but is that what you like? That is probably something you have to evaluate for yourself.
Why exactly are you building a tube amp?
 
Well to answer the question as to why i am building a tube amp, well that has a lot of answers that all play together:

-I like the look of Tubes, i hallways have (my grand parents gave me some old radio tubes when i was 5, and i have loved tubes ever since. And no i did not break them, i cherish those tube, i still have them all, and all but one survived the trip from Germany to the US and back to Germany again (in the cores of about 6 years)). My grandparents have a all tube amp, and it looks and sound good. (they don't use it much, now that they have a solid state amp as well)
- I like the sound of tubes, i have only listened to OT amps so far, but i have heard excellent things about OTL's
- I find that my transistor amp sounds to boring with my T1.
- I want something us full to build.

Almost every time im at my Grandparents house i listen to a Old Tube radio from the circa the 50 -60's. I do like the sound, but it could be a tiny mit less distorted though. But that may just be some components wearing down, it hasn't been serviced in about 35 years.
Now that i think of it, i do believe that the Radio uses a ECL86 as the audio output. I will check that in the schematics, as soon as i have time. (haven't looked into it for a while). I may be entirely wrong about that though.

edit:
The sound of my T1 isn't actually boring, in most songs, but in some it it is, but i still want the sound of tubes OTL or OT. Heck i might just build both designs and compare the sound. As id use the same tube in the 2 designs. Id build the OTL first, as it is smaller and less expensive. And all the recommended amps for the T1 are OTL...
 
Last edited:
now the only questions that remain at this point in time, is which of these schematics i should use as a base?


And what i should change / add. The power supply i intend to use will be similar to the one in this schematic: http://www.iol.ie/~waltonaudio/pcl86images/pcl86schematicbig2mod.jpg

And the last question in this post is, recommended brands for the components,
for the low voltage side, i will most likely use Elna Silmic II caps.
I don't know a good company for the higher voltage stuff. As well as the resistors. As well as Power Transformer.

I will be using this power transformer: TA_PCL86

I will controll the low voltage DC current using a semiconducter voltage regulator to 12VDC (i asume i the tube can handle a lower heater voltage?) If not i will see if i can build a system to get 13V exactly. Or ill have a custom transformer made for this amp.

Or if i use the ECL86 as apose to the PCL86 i will use this transformer: http://www.enovaz.it/prodotto-142825/TA002.aspx and no rectificaton.

Its just a question of, can i get a matched pair of them, and should i rectifier the heater current.
 
Last edited:
now the only questions that remain at this point in time, is which of these schematics i should use as a base?
These schematics were the first to popup when I made a quick search for examples, meaning it is probably possible to find something better or more suitable.

Maybe a kind member of this forum has something of the sort in his/her archives?

Anyway, they are a good starting point, and represent two different, not to say opposing philosophies.

The first one is rather straightforward and uses the pentode as a follower, which seems sensible. The gain is probably going to be much too large for this kind of application. A good way to manage the excess gain and improve the characteristics is to use NFB; it will mean a non-infinite input impedance, because of the configuration itself and the necessity to keep consistency over any source impedance, from zero to infinity, but with tubes, this is not difficult to manage, and inverting configurations tend to sound good.


The second example is more sophisticated, and already uses feedback (a lot) to control the output impedance: the natural impedance of the anode output stage is barely lower than the DC load resistors which is much higher than even 600ohm headphones.
It wouldn't be my first choice, since one asks the tube to do "unnatural things", but who knows: maybe it sounds alright. It seems to be competently designed anyway, and a current-driven load for open-loop conditions might be beneficial.



And the last question in this post is, recommended brands for the components,
for the low voltage side, i will most likely use Elna Silmic II caps.
I don't know a good company for the higher voltage stuff. As well as the resistors. As well as Power Transformer.
A good circuit doesn't require fancy components to work properly: use plain quality items, Elna is certainly more than OK, the risk if you go for highly sought-after brands is that you might end up with fakes and that is certainly a problem, much more than if you use just any cheap cap that does its job unglamorously, but properly.
You can validate your circuit using commodity components, and then upgrade when you're happy and you feel the need.
I will be using this power transformer: TA_PCL86

I will controll the low voltage DC current using a semiconducter voltage regulator to 12VDC (i asume i the tube can handle a lower heater voltage?) If not i will see if i can build a system to get 13V exactly. Or ill have a custom transformer made for this amp.

Or if i use the ECL86 as apose to the PCL86 i will use this transformer: TA002 and no rectificaton.

Its just a question of, can i get a matched pair of them, and should i rectifier the heater current.
At first sight, your transformers look OK. For an amplifier, you should not need DC heaters, the P-series in particular can cope with AC without problem.

I don't see what you mean by matched pair: tubes? transformers?
 
Last edited:
The big disadvantage of OTL with tubes is often a high output impedance. It really messes up the frequency responses of many headphones. You really want to keep it as low as you can. Some headphones (beyer mostly) are voiced for 120R outputs but they aren't the norm.

I'd use a derivation of the first schematic with, as suggested by Elvee, nfb added.

It probably would be quite easy to adapt Wavebourn Kaidan (made for the pcl84) for the pcl86: www.wavebourn.com • View topic - Wavebourn Kaidan vacuum tube headphone amplifier
 
A simple suggestion attached.

Btw the TA002 transformer is better in this case : 150ma B+ current vs 100ma for the TA_PCL86. With a 50% derating for class A and a cap input PS, you can get 35-40mA per channel with the first, barely 25ma with the second.
 

Attachments

  • pcl86.GIF
    pcl86.GIF
    11.3 KB · Views: 330
thedoci said:
Almost every time im at my Grandparents house i listen to a Old Tube radio from the circa the 50 -60's. I do like the sound, but it could be a tiny mit less distorted though. But that may just be some components wearing down, it hasn't been serviced in about 35 years.
The nearest thing in modern audio to a typical "Old Tube Radio" could be a tube guitar practice amp: single-ended pentode output, little or no negative feedback, deliberately rolled off HF response, almost random loudspeaker characteristics with boomy bass. Don't expect a decent hi-fi amp to sound like that.
 
Tanks for the inso TheGimp!
Would the amp benefit from a more soficticated desighn, as apose to the more simplistic one mentioned by member 00940?
Or is it better to go for a simple design.

Also a question the the design by 00940, i find it odd that there are no coupling caps in front of the tube grids. I have yet to see this design. What are the disadvantages and advantages of this design?
 
Tanks for the inso TheGimp!
Would the amp benefit from a more soficticated desighn, as apose to the more simplistic one mentioned by member 00940?
Or is it better to go for a simple design.
You can begin with a simple design: in fact, 00940 more or less materialized what I had in in mind: triode for the gain, CF as a follower and NFB to dump the excess gain in a useful way.
It is certainly possible to come up with something more clever and sophisticated (although none comes to my mind presently), but the basics would remain the same: the pentode has to be OP device, unless you are seriously out of of it, and it should preferably have a naturally low OP impedance, meaning CF (in principle: the second link I gave shows a "fabricated" low Z).
This means that the signal flow and power circuits will follow the same template, whatever fine detail you chose to implement.
Once you have built your amp, you can still make minor mods to make it more clever, but the base looks sound and healthy.
Also a question the the design by 00940, i find it odd that there are no coupling caps in front of the tube grids. I have yet to see this design. What are the disadvantages and advantages of this design
Regarding the first tube, he probably thought burdening the schematic with obvious details (a shunt resistor is also necessary) wasn't useful; the two tube sections are direct-coupled, which has the obvious drawback of dissipation, but with global feedback applied, eliminating one pole in the loop is a useful step towards preventing (or eliminating) motor-boating
 
Regarding the first tube, he probably thought burdening the schematic with obvious details (a shunt resistor is also necessary) wasn't useful;

the two tube sections are direct-coupled, which has the obvious drawback of dissipation, but with global feedback applied, eliminating one pole in the loop is a useful step towards preventing (or eliminating) motor-boating
Yes, I was a bit lazy and some stuff isn't needed in simulators (but very useful in real life). The pentode needs a grid stopper. In practice, I'd use a 10K potentiometer and no input caps.

There are some obvious improvements possible. You could replace the cathode resistor of the input triode by a diode (a 1N4148 should be fine). If the only headphones driven will be your 600ohms headphones, you can reduce the cathode follower current to about 25ma by increasing the resistor to ground to 5K6 (in which case you can use the ta_pcl86 transformer) and decrease the output capacitor size to 47uF or so. You can also increase gain by increasing the feedback resistor to 270K (gain of about 5).

If you wanted to have a more universal amp, I'd use two pcl86 per channel, with a White cathode follower at the output. I might post a possible schematic later on.
 
Last edited:
Push Pull in a headphone amp, that sounds very interesting. Also i intend to drive the fallowing impedance range: 25 to 600 ohm (and the 25 is the absolute bottom i would go, however for that headphone i think i will try and design a KT66 or KT88 based amp (but that is still about 4 years in the future). Its just future proofing the amp. A friend of mine also asked me if he can test the amp using his 32ohm headphones. So designing the amp with that range seems like a good idea. And based on what has been posted about OTL here before, it should be possible.

I also don't listen to music very loudly (the before mentioned 85 Db seem a fair bit too high to me, as 85Db makes my ears hurt after about 5 seconds of exposure, id guess i am listening to about 65 to 70 Db), and i will use a 150mA transformer. Also the transformer isnt an issue for me, i have no problem getting a big and heavy transformer in this amp. Although i have a upper limit in terms of weight. My desk can only support about 15KG in the area of the amp's current case design. My desk is made of glass, if your wondering.

Also Thanks for the schematics! And i am looking foreword to the PP OTL Schematic.

Thanks for the help, all of you guys!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.