Pelatuk OIPC
I have many variation of this topology, and I called it Pelatuk (a bird name) amplifier. Now, I want to change the compensation with OITPC. The goal of this amplifier is cheap enough, use transistors that easily to find in my country, and simple enough.
I have many variation of this topology, and I called it Pelatuk (a bird name) amplifier. Now, I want to change the compensation with OITPC. The goal of this amplifier is cheap enough, use transistors that easily to find in my country, and simple enough.
Attachments
This is PCB layout but not tested.
Please compare with simulation file (sometime I made mistake when convert schematic from simulation file to PCB layout software).
I want less distortion in high frequency. In the latest simulation, I use 2 sets of vertical mosfet with 45VDC PSU. It reduce high frequency distortion slightly and increase the slew rate.
I sim symmetry topology with cascode LTP dan Hawksford cascode VAS, which is very common design 
and output TEF. All BJT. Compensation is OITPC. It is very difficult to implement OITPC using output BJT rather than mosfet. I use 90VDC PSU.
This is loop gain and THD 20.
and output TEF. All BJT. Compensation is OITPC. It is very difficult to implement OITPC using output BJT rather than mosfet. I use 90VDC PSU. This is loop gain and THD 20.
Attachments
I sim symmetry topology with cascode LTP dan Hawksford cascode VAS, which is very common design
This is loop gain and THD 20.
Yes, triple BJT OPS introduces more phase shift (and harder to stabilize)than simple mosfet OPS.
According to my friend, he like the sound, specially the bass (many Indonesia people love the quality of the bass).
I have 4 pcs PCB, given by my other friend, but I have not build it, yet.
Hi Damir, sorry I am a late arriver to this thread.
Is the basic idea of OITPC the following:
1.- Start with regular 2 pole compensation (TPC)
2.- Then, add a capacitor from the T node in the 2 pole compensation to the output to make it OITPC?
Also, in TPC (referring to the schematics in post 1), there is no resistor R22 and R30. What is the purpose of these?
Thanks, Sandro
Is the basic idea of OITPC the following:
1.- Start with regular 2 pole compensation (TPC)
2.- Then, add a capacitor from the T node in the 2 pole compensation to the output to make it OITPC?
Also, in TPC (referring to the schematics in post 1), there is no resistor R22 and R30. What is the purpose of these?
Thanks, Sandro
https://www.diyaudio.com/forums/solid-state/338815-anistardi-peletuk-4.html#post5820836
Post 106 is not Pelatuk amplifier.
Last edited:
Hi Sandro,
Yes, the basic idea is like that, with some additional tweaking like capacitor parallel to the VAS emitter resistor and added pole to the one TPC leg.
Resistors R22 and R30 (they are actually in parallel) improves, by adding pol, the Phase and Gain margin all at higher frequencies than ULGF (ULGF not changed by this). This pol is used in ordinary TPC or TMC too.
All the best, Damir
Hi Damir,
I went through the fun process today to work out the TPC transfer functions, and yes resistors R22 and R30 add a pole to the network, but a zero to the amplifier to improve phase and gain margin.
Also, the more I stare at OITPC, the technique is an extension of the split Miller cap method. As you correctly pointed out, pure output inclusive Miller is impossible, your amp will go unstable. So you can do 2 things:
- TMC, or,
- Split the Miller cap by feeding some to the amp output and some to the VAS output (op-amps use this technique rather than TMC).
OITPC is an extension of the second one, where the cap that typically goes to the VAS output has been split, and the split portion is now fed to the output. That's how I think about OITPC now that it has sunk in.
Cool technique, simple and gives good results. Thanks for sharing. I
Best, Sandro
I went through the fun process today to work out the TPC transfer functions, and yes resistors R22 and R30 add a pole to the network, but a zero to the amplifier to improve phase and gain margin.
Also, the more I stare at OITPC, the technique is an extension of the split Miller cap method. As you correctly pointed out, pure output inclusive Miller is impossible, your amp will go unstable. So you can do 2 things:
- TMC, or,
- Split the Miller cap by feeding some to the amp output and some to the VAS output (op-amps use this technique rather than TMC).
OITPC is an extension of the second one, where the cap that typically goes to the VAS output has been split, and the split portion is now fed to the output. That's how I think about OITPC now that it has sunk in.
Cool technique, simple and gives good results. Thanks for sharing. I
Best, Sandro
Hi Damir,
This is really an excellent design.
See images below for OITPC and single pole (only 2*250pF from VAS, all other caps removed).
For some reason this .asc file works as expected, while the previous did not ?
As can be seen below, settling time is even shorter with no drama at all, nicely coming to rest as with the Single pole, but then faster.
My compliments for this design.
Hans
This is really an excellent design.
See images below for OITPC and single pole (only 2*250pF from VAS, all other caps removed).
For some reason this .asc file works as expected, while the previous did not ?
As can be seen below, settling time is even shorter with no drama at all, nicely coming to rest as with the Single pole, but then faster.
My compliments for this design.
Hans
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.