Hi Andrew
PSU freq +/- 33kHz
Could this be ripple that is amplified ?
The PSU ripple measures +/- 76mV rms
PSU freq +/- 33kHz
Could this be ripple that is amplified ?
The PSU ripple measures +/- 76mV rms
The main HF ripple is being modulated by a lower frequency. I wonder if this modulating frequency is the 33kHz PSU leaking through?
If it is, then the PSU frequency is passing the filters on the PSU output AND the amplifier PSRR is still letting some get to the amp output.
Could the PSU be sending the amp into an oscillation?
If it is, then the PSU frequency is passing the filters on the PSU output AND the amplifier PSRR is still letting some get to the amp output.
Could the PSU be sending the amp into an oscillation?
If I probe the transconductance stage that waveform is so minute that on the lowest scale it can barely be seen. That waveform is present at both sides of the bias transistor.
Strange thing is that Mike's original amp layout ver 5.1 was tested with an enclosed switching psu (that I had built a while ago) that has the exact same switching freq. and that troublesome waveform is non existant in the VAS 😕
Strange thing is that Mike's original amp layout ver 5.1 was tested with an enclosed switching psu (that I had built a while ago) that has the exact same switching freq. and that troublesome waveform is non existant in the VAS 😕
Attachments
Symasym5.3 stuffed PC board on the Ebay
HI
I'm seling my Symasym 5.3 amplifier stuffed PC boards on Ebay .
I used exotic parts like WBT solder Holco/PRP resistor etc you can read more at Ebay # 280198567358 .
It is realy a great amp but I just moved a new apartment so I desided I keep only one amp . All the rest must go . Way to big problem when I have 15amp preamp etc and I'm moving .
Regards
HI
I'm seling my Symasym 5.3 amplifier stuffed PC boards on Ebay .
I used exotic parts like WBT solder Holco/PRP resistor etc you can read more at Ebay # 280198567358 .
It is realy a great amp but I just moved a new apartment so I desided I keep only one amp . All the rest must go . Way to big problem when I have 15amp preamp etc and I'm moving .
Regards
Help unable to bias my symasym 5.3
Hi Folks,
I'm having difficulty setting bias on my symasym 5.3
the bias stays at 0 mA.
I made my own pcb based on the eagle schematics ,
I've double checked my PCB according to the schematics,
no problem there.
i'm bit puzzled with the eagle schematics and the complte sch.
there're adirference on the VBE and feedback, which one is the right one?
I've checked all the parts, solder koint and all seems good.
also changed the MJE 15030/31 with 2SC 4793/SA 1837
Please folks need some advice
many thanks in advance
best rgds.
Hi Folks,
I'm having difficulty setting bias on my symasym 5.3
the bias stays at 0 mA.
I made my own pcb based on the eagle schematics ,
I've double checked my PCB according to the schematics,
no problem there.
i'm bit puzzled with the eagle schematics and the complte sch.
there're adirference on the VBE and feedback, which one is the right one?
I've checked all the parts, solder koint and all seems good.
also changed the MJE 15030/31 with 2SC 4793/SA 1837
Please folks need some advice
many thanks in advance
best rgds.
Hi Dexter, sorry for the late response...
I guess it's time for some measuring. On my webpage you find at the bottom a diagram with "should be" currents.
You should start measuring the voltage across r5/6, it should be each ~1v, or at least the sum needs to be 2v. R10 should read ~0.4v. Next check r15/17, they should have ~0.37v.
Tell the measured voltages, this should help locating the problem.
The differences between the schematics are some swapped parts that are in series, this does not make any change.
What was wrong about my pcb?
Mike
I guess it's time for some measuring. On my webpage you find at the bottom a diagram with "should be" currents.
You should start measuring the voltage across r5/6, it should be each ~1v, or at least the sum needs to be 2v. R10 should read ~0.4v. Next check r15/17, they should have ~0.37v.
Tell the measured voltages, this should help locating the problem.
The differences between the schematics are some swapped parts that are in series, this does not make any change.
What was wrong about my pcb?
Mike
Many Thanks Mike for your reply.
Your Pcb artwork is excellent.
I made my own pcb because I want to add the psu onboard
for less cabling mess just like AAK's pcb.
I'll check the voltage on those R's and I'll get back to you asap.
many thanks and best rgds,
Your Pcb artwork is excellent.
I made my own pcb because I want to add the psu onboard
for less cabling mess just like AAK's pcb.
I'll check the voltage on those R's and I'll get back to you asap.
many thanks and best rgds,
Mike,
I measured the following
r5/6: 1.001 and 1.001
R10: 0.347V.
R15/17: 0.367V and 0.385V
I'll calculate the mA into mV and strat probing
Can you suggest what other things I should measure?
The Dc offset is 0.1mV for both channel
Thak you for your help
best rgds,
I measured the following
r5/6: 1.001 and 1.001
R10: 0.347V.
R15/17: 0.367V and 0.385V
I'll calculate the mA into mV and strat probing
Can you suggest what other things I should measure?
The Dc offset is 0.1mV for both channel
Thak you for your help
best rgds,
Hi dexter, your measured values are perfectly fine. This means input stage and vas are functional. So, the problem is in the outputstage / vbe-multiplier.
Try measuring the voltage T1: C -> E, base U5 to base U6.
Mike
Try measuring the voltage T1: C -> E, base U5 to base U6.
Mike
HI mIke,
Many thanks for the reply.
I measured 1.5V between U5 and U6.
thanks mike and best rgds,
Many thanks for the reply.
I measured 1.5V between U5 and U6.
thanks mike and best rgds,
Can you adjust these 1.5v up to 2.4v with the bias-pot? Have you accidently reversed the bd139 (T1)? It has e-c-b pinout.
@ MikeB
in my symasym
the DC offset is still around 7mV for both channel, even I have change the differential with match hFE (I use cheap multimeter to match it). Is that OK?
And I use toshiba for output transistor. Have you try to put c5200 and a1943 with same parts (refer to version 5.3)?
only feedback caps is change from 22pf to 10pf
how to know if my symasym is oscilated? I don't have stuff like osciloscope
sorry for my BAD english, I hope you understand what i mean.
Thank you
in my symasym
the DC offset is still around 7mV for both channel, even I have change the differential with match hFE (I use cheap multimeter to match it). Is that OK?
And I use toshiba for output transistor. Have you try to put c5200 and a1943 with same parts (refer to version 5.3)?
only feedback caps is change from 22pf to 10pf
how to know if my symasym is oscilated? I don't have stuff like osciloscope
sorry for my BAD english, I hope you understand what i mean.
Thank you
Mike,
thanks for the reply.
I turned the pot and the voltage stays at 1.5V.
for bd139 vbe, I think the pins're correctly placed
I've checked all the tr w/ my dmm and they all seems ok.
best rgds,
thanks for the reply.
I turned the pot and the voltage stays at 1.5V.
for bd139 vbe, I think the pins're correctly placed
I've checked all the tr w/ my dmm and they all seems ok.
best rgds,
no.crt said:
There are three pairs of transistors that must be matched.
Using a DMM hFE tester with unknown base or collector current is not good enough.
You must match each of the pairs at the operational Ic as in the working circuit.
I also match the Vbe at the operational current. In fact I consider Vbe matching to be far more important than hFE in the matching.
Hi Andrew,
If you match transistors for beta from the same batch, the Vbe will also be matched. The reverse is not true. If you are trying to match parts from different batches .... why? It really is that simple.
Now, if you can match beta along several different emitter currents (close to collector currents if you want to get picky), then the match of Vbe will only affect DC offset by a small amount. The beta matching is crucial for AC performance as this is your error corrector. The most critical of all things to match. Try it with a good THD meter.
We really do agree on most things.
Hi crt,
Andrew is dead right on disagreeing with your matching method. Device temperature will greatly affect your readings. I designed a jig, and I'm sure most of us have "designed" that same jig. Basically, you build a differential pair with sockets for your two parts. Let the temperature settle and have the two parts touching each other. The ones that pass equal current are matched. I measure across collector resistors. You need matched resistors for the two base resistors (to common ground) and your two collector load resistors. Mine has variable current, but you can fix yours to 2 mA (1 mA per transistor) to match reasonably closely most situations. I also generally run collector voltages of 10 VDC. So this is easy and inexpensive to build.
-Chris
If you match transistors for beta from the same batch, the Vbe will also be matched. The reverse is not true. If you are trying to match parts from different batches .... why? It really is that simple.
Now, if you can match beta along several different emitter currents (close to collector currents if you want to get picky), then the match of Vbe will only affect DC offset by a small amount. The beta matching is crucial for AC performance as this is your error corrector. The most critical of all things to match. Try it with a good THD meter.
We really do agree on most things.
Hi crt,
Andrew is dead right on disagreeing with your matching method. Device temperature will greatly affect your readings. I designed a jig, and I'm sure most of us have "designed" that same jig. Basically, you build a differential pair with sockets for your two parts. Let the temperature settle and have the two parts touching each other. The ones that pass equal current are matched. I measure across collector resistors. You need matched resistors for the two base resistors (to common ground) and your two collector load resistors. Mine has variable current, but you can fix yours to 2 mA (1 mA per transistor) to match reasonably closely most situations. I also generally run collector voltages of 10 VDC. So this is easy and inexpensive to build.
-Chris
I have found that matching hFE does not automatically give good or any match for Vbe. But maybe my group of one or two hundred are not from the same batch.anatech said:
As a result of this finding, I match Vbe as close as I can manage using your jig which ensures matched Vbe and check for equal Ic in the balanced pair. I then read off the Ib from the base resistor voltage drop and try to get those within 5%.
I then go back to the identified possible pairs and remeasure with a zero value base resistor (temp short across the connections) to recheck the Ic values are within 1%. Many possible pairs fail this last test.