Hi!
Before testing new 5r6 values I measured voltage drop between transistor bases and it was around 1v. The Vbe out of circuit are 0.6v and 0.630v. I measured the transistors before I installed them in.
I tried 10r and 15r without luck. And then finally 20r. This last value worked, the distortion is gone
But another issue comes with that. The transistors are hotter that before. I was adjusting the trimmer for the half supply and noticed that it was around 15v instead of 17v. I turned the trimmer and the voltage didn't change much. Then I noticed how the supply voltage was going down to 30v and below and then the 250mA fuse blew out.
(note: supply voltage has always been between 33v an 34v)
15r didn't work and 20r did. But maybe the 20r was too much and the transistors were drawing too much current?
God, this is the never ending amp.
Before testing new 5r6 values I measured voltage drop between transistor bases and it was around 1v. The Vbe out of circuit are 0.6v and 0.630v. I measured the transistors before I installed them in.
I tried 10r and 15r without luck. And then finally 20r. This last value worked, the distortion is gone
But another issue comes with that. The transistors are hotter that before. I was adjusting the trimmer for the half supply and noticed that it was around 15v instead of 17v. I turned the trimmer and the voltage didn't change much. Then I noticed how the supply voltage was going down to 30v and below and then the 250mA fuse blew out.
(note: supply voltage has always been between 33v an 34v)
15r didn't work and 20r did. But maybe the 20r was too much and the transistors were drawing too much current?
God, this is the never ending amp.
Pay attention to # 10. Use the most affordable darlington pairs. You will have to organize Ube in the bias chain (#79).
Perhaps there is self-excitation. Increase the 22p capacitor in steps to 100p. A complete examination requires an oscilloscope or sound card and audio software.
You can use class A if you have large heatsinks and a powerful power supply unit (as in JLH1969))).
JLH 10 Watt class A amplifier
Perhaps there is self-excitation. Increase the 22p capacitor in steps to 100p. A complete examination requires an oscilloscope or sound card and audio software.
You can use class A if you have large heatsinks and a powerful power supply unit (as in JLH1969))).
JLH 10 Watt class A amplifier
Last edited:
At a voltage of 32V, you can use PCB amplifiers on the TDA2030 with a monopolar power supply. The most basic Chinese.
https://aliexpress.ru/item/32899760...h3DnwXqEAEYASABEgKT3fD_BwE&sku_id=65806662855
https://aliexpress.ru/item/32899760...h3DnwXqEAEYASABEgKT3fD_BwE&sku_id=65806662855
Last edited:
The pot has a resistance of approximately 1–2 kΩ.
Transistor any TO-92 or TO-126 (polarity in accordance with the conductivity npn or pnp). When you first turn on the central (movable) contact to the collector (minimum current). Turn, move, control the current consumption. TO-126 can be installed together with output transistors for better control of heating at the installation site.
схемы смещения транзисторных выходных каскадов - Google Search
Transistor any TO-92 or TO-126 (polarity in accordance with the conductivity npn or pnp). When you first turn on the central (movable) contact to the collector (minimum current). Turn, move, control the current consumption. TO-126 can be installed together with output transistors for better control of heating at the installation site.
схемы смещения транзисторных выходных каскадов - Google Search
Last edited:
Good news! I'm still with the original diode+resistor biasing thing. It turns out that 16.5r is perfect. I tried with the songs that sounded awful and I don't hear any distortion. The output transistors are warm but not really hot. I can't believe how a difference of 1r5 can do that much. With 15r the distortion was very obvious.
Voltage drop between bases is around 1.25v. Has the voltage drop between bases something to do with the actual Vbe of the transistor used? I mean, one of the transistors has Vbe = 0.6v and the other one Vbe = 0.64v. If we do the maths its 1.24v. Does it mean that I need at least a voltage drop of 1.24v between bases? If so, I would say that a tiny bit more just to be safe, but not too much because of insane current consumption.
Thanks, I'm learning a lot
Voltage drop between bases is around 1.25v. Has the voltage drop between bases something to do with the actual Vbe of the transistor used? I mean, one of the transistors has Vbe = 0.6v and the other one Vbe = 0.64v. If we do the maths its 1.24v. Does it mean that I need at least a voltage drop of 1.24v between bases? If so, I would say that a tiny bit more just to be safe, but not too much because of insane current consumption.
Thanks, I'm learning a lot
There is still a drop of 0.5 (1r). Together we get 1.25v.
Distortion pattern of your amplifier from the 1966 manual.
Harmonic Distortion: @ 1 KHz @ 8 ohms
0.1W 0.25%
0.5W 0.55%
1.0W 0.95%
1.5W 2.2%
10W 10%
That's a lot (by modern standards), but the ceramic phono head distorts more.
Distortion pattern of your amplifier from the 1966 manual.
Harmonic Distortion: @ 1 KHz @ 8 ohms
0.1W 0.25%
0.5W 0.55%
1.0W 0.95%
1.5W 2.2%
10W 10%
That's a lot (by modern standards), but the ceramic phono head distorts more.
Last edited:
It is normal, if the bias diode is not thermally connected to the output transistors!
Remember the Vbe characteristic: ~60mV / decade-Collector-current.
You're experiencing thermal runaway -- and it's a risk to any bipolar design that handles enough power to warm up appreciably!
Have another look at OldDIY's 'bias spreader' (post 79). You can achieve better results than you're getting, and with just about any transistor (mounted on the heatsink near the outputs), and with little consideration given the resistor values. A 1k trimpot with an 820 ohm Base to Collector would work. If the transistor has an hFE in the 150 range or better, 10k and 8k2 would work; even 25k and 18k if it's still higher.
Cheers
Remember the Vbe characteristic: ~60mV / decade-Collector-current.
You're experiencing thermal runaway -- and it's a risk to any bipolar design that handles enough power to warm up appreciably!
Have another look at OldDIY's 'bias spreader' (post 79). You can achieve better results than you're getting, and with just about any transistor (mounted on the heatsink near the outputs), and with little consideration given the resistor values. A 1k trimpot with an 820 ohm Base to Collector would work. If the transistor has an hFE in the 150 range or better, 10k and 8k2 would work; even 25k and 18k if it's still higher.
Cheers
Last edited:
Try to keep in mind, while you're working on this piece (and others like it), that the designer's *prime directive* was, 1st) eliminate every possible $0,79 part, then 2nd) eliminate every possible $0,49 part, then 3rd) eliminate every possible $0,19 part, and so on ..
The guys that designed this sort of gear got their training when valves were king, and regarded every component as both an initial cost to be minimized, and a future source of failure.
Just something to keep in mind when you spot a design like this one -- with obsolete practices scattered about.
Regards
The guys that designed this sort of gear got their training when valves were king, and regarded every component as both an initial cost to be minimized, and a future source of failure.
Just something to keep in mind when you spot a design like this one -- with obsolete practices scattered about.
Regards
1.Repeat the bias setting again immediately after working at full volume (heating).
2. Move the diode to the place where the output transistors are installed. Provide thermal contact (thermal grease). You can replace the diode with a TO-126 transistor with connected BCs. Connect wires to PCB.
3. You can try two diodes: 2x5148 or different types. See point 2. An adjustment resistor can be connected in parallel to one of the diodes.
4. Add 'bias spreader' (post 79) between the bases of the output transistors. The transistor must have a thermal contact.
The original amplifier works in class B . Therefore, there is no thermal runaway. But the sound alas ...
2. Move the diode to the place where the output transistors are installed. Provide thermal contact (thermal grease). You can replace the diode with a TO-126 transistor with connected BCs. Connect wires to PCB.
3. You can try two diodes: 2x5148 or different types. See point 2. An adjustment resistor can be connected in parallel to one of the diodes.
4. Add 'bias spreader' (post 79) between the bases of the output transistors. The transistor must have a thermal contact.
The original amplifier works in class B . Therefore, there is no thermal runaway. But the sound alas ...
Last edited:
I'm trying the bias spreader with a bd139 between the output tranies at the heatsink. For a NPN (bd139), the emitter goes to the base of the PNP output transistor?
Thanks for the resistor values Rick! Let's hope it works, it's the last step
About the design of this thing...God, it's hard to believe how that thing could even work at some time in the past without some personal adjustemet.
Many thanks!
Thanks for the resistor values Rick! Let's hope it works, it's the last step
About the design of this thing...God, it's hard to believe how that thing could even work at some time in the past without some personal adjustemet.
Many thanks!
Last edited:
I've just tried the bias spreader from #79 with no luck. Used a 7.5k resistor, a 10k pot and a BD139 with the emitter to the PNP base transistor.
The little hum you hear through the speaker when the you turn the amp gets louder very fast after a couple seconds and I know what come next -> blown fuse. So I turn off the amp before that happens. It happens very fast. I tried a few positions of the trimmer, all of them with same result.
I don't know if that's related with unappropriated resistor+trimpot value or what. Damn.
The little hum you hear through the speaker when the you turn the amp gets louder very fast after a couple seconds and I know what come next -> blown fuse. So I turn off the amp before that happens. It happens very fast. I tried a few positions of the trimmer, all of them with same result.
I don't know if that's related with unappropriated resistor+trimpot value or what. Damn.
Take R1 1kom. P2 preferably 2-3kom (10kom may work). Connect the movable contact P2 to the top contact according to the diagram. Set resistance to maximum. Base T1 BD139 (left) is connected to the connection of resistors (middle). Emitter T1 (right) to the free contact of the pot P2 (bottom). The collector (center) T1 is connected to the free end of R1 (top).
T1 BD139 is installed on top of one of the output transistors under the same screw (thermal contact).
The whole structure is included between the bases according to scheme # 79 .
We include. Rotate the pot slowly (decrease). We control the offset. If everything is fine, let the amplifier work, and again control the offset. If necessary, we adjust (without sudden movements))).
BD139 can be used as a diode. Connect C-B (this is +), E -
T1 BD139 is installed on top of one of the output transistors under the same screw (thermal contact).
The whole structure is included between the bases according to scheme # 79 .
We include. Rotate the pot slowly (decrease). We control the offset. If everything is fine, let the amplifier work, and again control the offset. If necessary, we adjust (without sudden movements))).
BD139 can be used as a diode. Connect C-B (this is +), E -
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.