Hi
Has Anyone done this to any tube amps,Are there Caps useds?
Thanks for your time
Install a "snubber circuit" consisting of a 10ohm - 10W resistor (2 resistors actually) at the two wires from the input transformer to the power supply circuit board. The resistors are soldered in line with the two wires before connecting to the power supply circuit board. Resistor values of 3ohm - 10ohm can be used, 10W rating is important. You will have to re-bias the tubes after this mod. The idea is to reduce AC noise before the rectifiers. A better snubber circuit utilizes an RC circuit. I have not experimented with caps in the snubber circuit yet.
Has Anyone done this to any tube amps,Are there Caps useds?
Thanks for your time
Install a "snubber circuit" consisting of a 10ohm - 10W resistor (2 resistors actually) at the two wires from the input transformer to the power supply circuit board. The resistors are soldered in line with the two wires before connecting to the power supply circuit board. Resistor values of 3ohm - 10ohm can be used, 10W rating is important. You will have to re-bias the tubes after this mod. The idea is to reduce AC noise before the rectifiers. A better snubber circuit utilizes an RC circuit. I have not experimented with caps in the snubber circuit yet.
This article gives a good overview on calculating RC snubbers:
www.hagtech.com/pdf/snubber.pdf
It's a fair bit of math, but the results are worth it.
www.hagtech.com/pdf/snubber.pdf
It's a fair bit of math, but the results are worth it.
Here is a method for the mathmatically handicapped (such as myself). Modified from a maxim application note
1. Measure the frequency of the spike resonance.
2. Add a shunt capacitor across the parasitic component and adjust the value of this capacitor until the frequency of the spike resonance is reduced by a factor of two. The value of this resulting capacitor will be three times the value of the parasitic capacitance that is creating the voltage spikes.
3. Because the parasitic capacitance is known, the parasitic inductance can be determined using the formula:
L = 1 / [(2 x Pi x F)² x C] where F = resonant frequency and C = parasitic capacitance
4. Now that both the parasitic capacitance and inductance are known, the characteristic impedance of the resonance can be determined using the following formula:
Z = SQRT(L/C) where L = parasitic inductance and C = parasitic capacitance
5. The resistor in the RC snubber circuit should be sized to the value of the characteristic impedance, and the capacitor should be sized between four and ten times the parasitic capacitance. The use of larger capacitors slightly reduces the voltage overshoot at the expense of greater power dissipation and less inverter efficiency.
1. Measure the frequency of the spike resonance.
2. Add a shunt capacitor across the parasitic component and adjust the value of this capacitor until the frequency of the spike resonance is reduced by a factor of two. The value of this resulting capacitor will be three times the value of the parasitic capacitance that is creating the voltage spikes.
3. Because the parasitic capacitance is known, the parasitic inductance can be determined using the formula:
L = 1 / [(2 x Pi x F)² x C] where F = resonant frequency and C = parasitic capacitance
4. Now that both the parasitic capacitance and inductance are known, the characteristic impedance of the resonance can be determined using the following formula:
Z = SQRT(L/C) where L = parasitic inductance and C = parasitic capacitance
5. The resistor in the RC snubber circuit should be sized to the value of the characteristic impedance, and the capacitor should be sized between four and ten times the parasitic capacitance. The use of larger capacitors slightly reduces the voltage overshoot at the expense of greater power dissipation and less inverter efficiency.
I agree. I built a regulated supply once and the damped oscillations (caused by diode reverse recovery exciting the parasitic resonant circuit in the power transformer secondary) slipped through the regulator without much attenuation. The right way to solve this problem is lowering Q, and a resistor offers easy control of Q. Caps just lower the frequency of the oscillation and attenuate a bit. A resistor stops the oscillations dead.
The fundamental reason for the switching transient is the current in the secondary winding inductance, which partially dissipates in to the other windings as well as charging the diode node voltage (ie. including the diode parasitic capacitance). The diode path is sometimes a relatively large loop.
An alternative method of managing the transient is to place capacitance across the secondary winding, with as 'small' a loop inductance as possible - which can be used to divert a significant portion of the energy in the winding leakage inductance. This bypass capacitance can also benefit from close to critical damping.
If high frequency noise is causing noticable problems, then this method can benefit from added special inductance in series with each diode - such as amorphous core amobeads.
Ciao, Tim
An alternative method of managing the transient is to place capacitance across the secondary winding, with as 'small' a loop inductance as possible - which can be used to divert a significant portion of the energy in the winding leakage inductance. This bypass capacitance can also benefit from close to critical damping.
If high frequency noise is causing noticable problems, then this method can benefit from added special inductance in series with each diode - such as amorphous core amobeads.
Ciao, Tim
Everything is a tradeoff. More shunt capacitance = lower transient voltage disturbance and lower ring frequency - but higher mains current.
If you have a problem, then that provides the best benchmark for choosing a value.
I happily add 10nf to HT secondaries - but that is just a gut feeling.
If you have a problem, then that provides the best benchmark for choosing a value.
I happily add 10nf to HT secondaries - but that is just a gut feeling.
Did this mod
Hi
I have two Small Tube mono amps that i got for $20 just play with,precision
electronice 30 watts,thay have this setup, 5ohm 5w an sound ok so i put two 6ohm 10w in my Jolida 502a just to see how it sounds.As with some other amp i have like Dyn MK3 with tube diodes, the 500v B+ is high i use a 3mf cap as the frist cap
before the chok were a 30mf was,this drops the B+ to 460 sounds much better i run the amp in pentoid not UL an run the output tube cathoids out to the 0 an the 16 ohm taps an move the Grond to the 4ohm tap LIke ARC,This gives more output dont see why others dont do this? I see other use lower B+ like ARC set up most of all there amps at 440 on the B+. lower B+ lets me bias the output tube more sounds sweeter.
I was thinking that i would try this Snubber mod on the MK3s.
Thanks for all the info,more is better as you all no.
Install a "snubber circuit" consisting of a 10ohm - 10W resistor (2 resistors actually) at the two wires from the input transformer to the power supply circuit board. The resistors are soldered in line with the two wires before connecting to the power supply circuit board. Resistor values of 3ohm - 10ohm can be used, 10W rating is important. You will have to re-bias the tubes after this mod. The idea is to reduce AC noise before the rectifiers. A better snubber circuit utilizes an RC circuit. I have not experimented with caps in the snubber circuit yet.
Hi
I have two Small Tube mono amps that i got for $20 just play with,precision
electronice 30 watts,thay have this setup, 5ohm 5w an sound ok so i put two 6ohm 10w in my Jolida 502a just to see how it sounds.As with some other amp i have like Dyn MK3 with tube diodes, the 500v B+ is high i use a 3mf cap as the frist cap
before the chok were a 30mf was,this drops the B+ to 460 sounds much better i run the amp in pentoid not UL an run the output tube cathoids out to the 0 an the 16 ohm taps an move the Grond to the 4ohm tap LIke ARC,This gives more output dont see why others dont do this? I see other use lower B+ like ARC set up most of all there amps at 440 on the B+. lower B+ lets me bias the output tube more sounds sweeter.
I was thinking that i would try this Snubber mod on the MK3s.
Thanks for all the info,more is better as you all no.
Install a "snubber circuit" consisting of a 10ohm - 10W resistor (2 resistors actually) at the two wires from the input transformer to the power supply circuit board. The resistors are soldered in line with the two wires before connecting to the power supply circuit board. Resistor values of 3ohm - 10ohm can be used, 10W rating is important. You will have to re-bias the tubes after this mod. The idea is to reduce AC noise before the rectifiers. A better snubber circuit utilizes an RC circuit. I have not experimented with caps in the snubber circuit yet.
Attachments
![!Bp9RydwBGk~$(KGrHqEOKjYEuc7bWo4+BLtm7z6FYw~~_12[1].JPG](/community/data/attachments/212/212213-70601bd7441ad96eeeb46bce06920fe1.jpg)
I'm with you, trobbins. My supplies are quite ring-free, but caps (even R-C) across the diodes did very little for me. It's all about taming the inductive kick from the transformer, and sometimes from the choke too.
My trial and error method has been to have a large assortment of high WVDC metallized film caps. I am preferential to the CDE 940C series, as there is ample stock and good selection of values. Zoom in on the turn on/off transient present at the transformer secondary and start with the lowest value cap, just a cap. If you see the frequency decrease, you are on the right track. If not, you need more capacitance. Once you get in the right neighborhood, start adding series resistance experimentally. I just put in a pot for starters and see how the range of 10 ohm to 1k affects the transient. If the resistance isn't enough to make the waveform nice and smooth, step up the capacitance and try again with the resistor.
Caution is needed to keep your fingers away from high voltages, and watch it that you don't blow your scope input.
I have had reasonable success snubbing the 120VAC winding instead, which allows lower VDC caps yet still keeps the secondary fairly quiet.
My trial and error method has been to have a large assortment of high WVDC metallized film caps. I am preferential to the CDE 940C series, as there is ample stock and good selection of values. Zoom in on the turn on/off transient present at the transformer secondary and start with the lowest value cap, just a cap. If you see the frequency decrease, you are on the right track. If not, you need more capacitance. Once you get in the right neighborhood, start adding series resistance experimentally. I just put in a pot for starters and see how the range of 10 ohm to 1k affects the transient. If the resistance isn't enough to make the waveform nice and smooth, step up the capacitance and try again with the resistor.
Caution is needed to keep your fingers away from high voltages, and watch it that you don't blow your scope input.
I have had reasonable success snubbing the 120VAC winding instead, which allows lower VDC caps yet still keeps the secondary fairly quiet.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.