Plh

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
V

VladimirK

Member
Joined 2007
johnthesoftware said:
Well, I didn't expect it to sound better than the JLH, just LOUDER, which it is, but the bass is ill-defined. There's a lot more of it at the lower end, which is fine, but it seems to swallow up all adjacent frequencies so, for instance, the ridiculously over-recorded double bass on the more recent Jacques Loussier records which sounds so fantastic on my JLH just sounds wooly on the PLH. Similarly Eminem (excellent recordings, who'd have thought).
Click to expand...

I beleive your point is correct, since I also assembled and listened to several similar amps, that differ mainly by a kind of output transistors.
Not speaking about overall sound, but only about naturality of bass, I could state, that bass gets better in this order
LMOSFET-VMOSFET-Si BJT-Ge BJT
Output impedance modulus does not say much about bass naturality.
For instance, 1mOhms Rout with lateral MOSFETS, achieved due to deep GNFB, produce nevertheless much more "gentle" and "plain" bass, than 100mOhms Rout with BJTs at No GNFB schematics.
 
V

VladimirK

Member
Joined 2007
flg said:
I'm not sure I undestand the cap thing. By making these series caps smaller you are high pass filtering. Cuttin off the bass! You might want to try more cap ;)
Click to expand...

I guess he meant the replacement 1x10000 by 10x1000
From my listenings, 10000 is not enough anyway (in spite that passband seems OK).
I would recommend, just try, adding 2x10000 more (possibly via 0,03...0,05 Ohms resistor, if these additional caps are lesser in quality than the first caps)
 
J

johnthesoftware

Member
Joined 2012
No, I did reduce the capacitance thereby high-pass filtering. It was my opinion that there was too much bass, so it was deliberate, and it increased the definition. It's as if there's a general resonance at lower frequencies.

Anyway, the point has become moot. The wind must have changed, I turned my back on them for half an hour and one of them fried.

I have become disillusioned and am going to the pub to talk about anything other than diy audio.
 
J

johnthesoftware

Member
Joined 2012
Ah ha, I think you're probably right Loudthud, I didn't think of that (this is mostly a learning exercise). A quick sim on both circuits shows the output impedance of the PLH to be about 2 and the JLH 0.16, a big difference.

Sadly there's not enough gain in the PLH to allow me increase the feedback. I would pursue this, but the fact that the quiescent current increases with temperature means either much bigger heat sinks or some sort of current control circuitry (much cheaper than the heatsinks), but it means repackaging the whole thing.

I'm going to consign this one to history and use my stock of FETs to drive my anti-gravity device.

Thanks to all for your help.

John
 
Loudthud

Loudthud

Member
Joined 2007
knutn

knutn

Member
Joined 2006
andiha.no
Paid Member
The DC offset was trimmed to be zero, and the variation afterwards was very small, less than 50 millivolts, if I remember right. The distortion was measured at 1 kHz with a 32 volt peak to peak output signal into eight ohms to be about -67 dB both second harmonic (2H) and third harmonic (3H) for the amplifier with the push pull output configuration. With single ended output was 2H: -49 dB, 3H: -58 dB, 4H: -70 dB and 5H: -79 dB. The distortion increased a little with an increase in the frequency. The idle current was set to 1.6 Ampere. With 2 Ampere, the distortion was lower. By now I am running the amplifier in push-pull mode with 1.6 Amperes idle current. The output impedance is measured to be about 0.4 ohms with this configuration. I am very pleased with this amplifier, but I think I would recommend a capacitance multiplier power supply, at least for the negative leg.
 
G

george_ct

Member
Joined 2011
Q1 from the PLH amplifier schematic does need a heat sink? How much power will dissipate?
 

Attachments

  • plh amplifier.png
    plh amplifier.png
    112 KB · Views: 542
diegomj1973

diegomj1973

Member
Joined 2010
2 picoDumbs said:
If you can modify circuit to use 2sj74 at input, and get rid of input cap, while maintaining same basic operation I would give that circuit a try.
Hard to beat Aleph J/J2/F6 circuits but this is also very nice.
Click to expand...

Where I live, unfortunately I do not get a variety of elements to be able to fly in my designs. I must limit myself to using very limited elements. It's a shame.

Anyway, I have many tools to evaluate with enough precision circuits, without having to mount anything, and get the projection of how this circuit will work once it is assembled. With feedback circuits, I rarely had to deviate too far from the projections, making value adjustments.

For example, with LabView and Multisim 14, I could literally hear how the schematic of an amplifier would sound, by importing a real .wav sound file and injecting it into its input. In the output of the schematic, I export the .wav file of how it would listen.

Likewise, I can emulate ripple noises and see how they are reflected in the output. Or I can analyze the crosstalk, for example, by giving the parameters of the actual wiring layout of the power supply to the amplifier channels.

With these tools, everything is limited by our imagination.

My greatest regards
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom