Hi,
At a first look at your drawing.
First Mu stage is not correct(Grid stopper is in the wrong postion)...should be the same as 2nd Mu stage.. Read this link..
Mu Stage Philosophy © 1993 Alan Kimmel
First tube is 6sn7 not SL7<<this was an early mistake on the thread..
I will have another look and say if I see any more problems..
IRF710 is a FET not as shown...
Driver stage is the technics style...bottom of PI to top tube.. OK I see you have swaped them over...
Anode of bottom tube?
NB the 150Ohm on the cathodes is only a guide it needs to be set based upon supply +/- voltage so I would start with 220 Ohm and reduce if required..Maybe even 300 Ohm..
PI resistors are 18K 2 Watt anode/cathode..
Value of R grid stopper 2nd stage Mu is 220 Ohm shown as CC type however It helps to fit a ferrite bead..
You could use a 115-115 V just the rails will be lower..
The rest looks OK at a first look...
PSU link was from here:
http://www.cincocoronas.com/nysva/images/25w_otl/25w_otl.pdf
However I think you will need to improve this to get hum lower on the rails...
Regards
M. Gregg
At a first look at your drawing.
First Mu stage is not correct(Grid stopper is in the wrong postion)...should be the same as 2nd Mu stage.. Read this link..
Mu Stage Philosophy © 1993 Alan Kimmel
First tube is 6sn7 not SL7<<this was an early mistake on the thread..
I will have another look and say if I see any more problems..
IRF710 is a FET not as shown...
Driver stage is the technics style...bottom of PI to top tube.. OK I see you have swaped them over...
Anode of bottom tube?
NB the 150Ohm on the cathodes is only a guide it needs to be set based upon supply +/- voltage so I would start with 220 Ohm and reduce if required..Maybe even 300 Ohm..
PI resistors are 18K 2 Watt anode/cathode..
Value of R grid stopper 2nd stage Mu is 220 Ohm shown as CC type however It helps to fit a ferrite bead..
You could use a 115-115 V just the rails will be lower..
The rest looks OK at a first look...
PSU link was from here:
http://www.cincocoronas.com/nysva/images/25w_otl/25w_otl.pdf
However I think you will need to improve this to get hum lower on the rails...
Regards
M. Gregg
Last edited:
Just for info..I decided that the Choke on the preamp PI was less important than the OP stage, so I removed the choke on the preamp stage and fitted a 10W 1.5K..So I now have 47uF 1.5K 47uF.<<no hum from pre stage..However this is also talking into account the twin Mu stage..
The home made chokes are now in the position of the preamp choke. This gave me more room to refit the single 1000uF 400V capacitor.
So the output power stage is; rectifier then single 1000uF 400V cap across the 150V +/- rails then choke in each leg +/- then 4700uF between each rail..This maintains the current drive for the Bass..
Russ andrews super clamp kit fitted to the mains...Looking at DC filters just for fun...
Hum is almost nothing...but there is always something more to improve..
Regards
M. Gregg
I've built plenty of amps that have no choke in the output supplies at all. Despite the sawtooth waveform, I get no hum in the speaker even with 107db efficiencies.
The real problem is not hum from the sawtooth, but diode commutation noise (switching noise). That can produce quite a buzz! But if the rectifiers in the power supplies are properly neutralized, that's all you need to sort that out.
The real problem is not hum from the sawtooth, but diode commutation noise (switching noise). That can produce quite a buzz! But if the rectifiers in the power supplies are properly neutralized, that's all you need to sort that out.
Hi,
I have snubbing on the diodes..However I have a few different values to try and see what I get..
I was looking at the toroids and wanted to see what if any difference a DC mains filter would make being an OTL...Have you tried DC filters?
Regards
M. Gregg
Just for interest..
Do you use Ceramic caps for snubbing ? Or the red wimas..I have seen 0.1uf used with 100 Ohm supposed to be better I have never tried them..
Anything that works any better with OTL topology ..Just thought I would ask...Thank's for any suggestions
Regards
M. Gregg
Do you use Ceramic caps for snubbing ? Or the red wimas..I have seen 0.1uf used with 100 Ohm supposed to be better I have never tried them..
Anything that works any better with OTL topology ..Just thought I would ask...Thank's for any suggestions
Regards
M. Gregg
Last edited:
Sometimes a toroid can get mechanically noisy from DC (asymmetry) on the mains and the DC filter can do a good job quieting them. So far not much difference in sound though.
The best way I have found to snub the diodes is to put a 'scope across the outputs and look at the noise floor (one must be careful to float the AC ground of the 'scope, BTW, else you will get erroneous distortion and power measurements, as the 'scope ground will short one of the speaker terminals to ground!). Then I put a network of a potentiometer and a cap across on the diode junctions and vary the pot to see if there is an effect. A 10K pot is plenty, you may have to play with cap values depending on the rectifier. Generally once I find a combo that works it will work on all the junctions. It takes time but does pay off, especially on super high efficiency (+105db) speakers. However, diode switching noise can show up even if the speaker is only 90 db.
The best way I have found to snub the diodes is to put a 'scope across the outputs and look at the noise floor (one must be careful to float the AC ground of the 'scope, BTW, else you will get erroneous distortion and power measurements, as the 'scope ground will short one of the speaker terminals to ground!). Then I put a network of a potentiometer and a cap across on the diode junctions and vary the pot to see if there is an effect. A 10K pot is plenty, you may have to play with cap values depending on the rectifier. Generally once I find a combo that works it will work on all the junctions. It takes time but does pay off, especially on super high efficiency (+105db) speakers. However, diode switching noise can show up even if the speaker is only 90 db.
Have you noticed this on Ultra fast soft recovery diodes / Hexfred?
Or is this mainly on standard diodes?
I'll have a play and see what if any gain I can get...Always looking for an improvement...
Regards
M. Gregg
Last edited:
on the Bob Pease (RIP) youtube lesson on scope probes one of the RF guys says that the first thing he does is cut off the ground pin so that the scope really floats. i hope he was saying this in jest, because the Tektronix guy showing off the 9GHz DSO probably had his eyes roll into the back of his head. everybody should own at least one differential probe!
the nice thing about the Tek passive probes is that the ground clip doesn't start glowing when you've a ground fault, part of the metal just vaporizes!
i've found that with most toroid transformers and diode junction capacitance that i work with the snubber cap is in the low nano-Farads, and the snubber resistor can be anywhere from a few hundred Ohms, to a few thousand. it's ok to parallel the snubber with 10 to 100nF ceramic cap.
The HEXFREDs are great in driver power supplies (no snubbing needed). In an output section they can be fragile- their rating can never be exceeded reliably (most Si rectifiers have a 10x surge rating) and its a good idea to snub them as well.
I'm going to try your ideas and I will look at the snubbing much closer..would you say that OTL amps are more sensitive to these issues?
Would Schottky diodes be better in the Op stage?
Regards
M. Gregg
Last edited:
OTLs are not more sensitive in particular but Circlotrons can be, depending on how the power supply is built and the application. In an OTL application there is an incentive to keep the series resistances of the power supply rail to a minimum, to make maximum current available. People don't think about this very much, but OTLs have to directly deliver the current that is part of the power that moves the speaker, so they can have a lot in common with transistor amps in that regard.
IOW, what works in a transistor amp for the output section power supply is how you may have to do it for an OTL. Transistor amps can be made quiet enough though. We've been able to get our amps to be silent on speakers that are 107 db without going through excessive means; grounding can have a lot to do with that too.
Schottky diodes would have to be snubbed too.
IOW, what works in a transistor amp for the output section power supply is how you may have to do it for an OTL. Transistor amps can be made quiet enough though. We've been able to get our amps to be silent on speakers that are 107 db without going through excessive means; grounding can have a lot to do with that too.
Schottky diodes would have to be snubbed too.
Last edited:
Just an Idea from me:
Should it be possible to use 6AS7G instead of the 6C33C?
I saw some builds with 8 of them and they are also 25W/4Ohm ones.
For example: 6AS7GA OTL
What resistor does i have to change? I think the 120 one to a 100 one?
Should it be possible to use 6AS7G instead of the 6C33C?
I saw some builds with 8 of them and they are also 25W/4Ohm ones.
For example: 6AS7GA OTL
What resistor does i have to change? I think the 120 one to a 100 one?
Forget this, i need some sleep
This would be 8 tubes per channel what would cost more, what is bad )=
Forget this, i need some sleep
This would be 8 tubes per channel what would cost more, what is bad )=
This circuit was discussed a while back in this thread..
Take a look at the PI PSU voltage (This is a way of getting more drive) and pre stage voltages..
I guess you could drive this OP stage, however it's probaby better to just build it as is..In discussion some on here have built it and say it sounds very good!
The Idea of autobias may be a problem and the heater voltages required..
I think it would be a lot of effort for autobias lower power..so its probably better to just build it fixed bias..
Regards
M. Gregg
I think,
Its important to do alot or research before you build anything..
I don't think OTL is difficult, however you have to look at the heat from OP tubes power used and power out that you want from a project.
The 6C33C is a very hot tube..and power just for heaters is high.
Other tubes are used in parallel to get the same power...again you have to add up the heater current the gain may be in heat spread over a larger area..
Here is an example
http://www.transcendentsound.com/Transcendent/Products.html
I have no connection to this link just for information..
Its important to do alot or research before you build anything..
I don't think OTL is difficult, however you have to look at the heat from OP tubes power used and power out that you want from a project.
The 6C33C is a very hot tube..and power just for heaters is high.
Other tubes are used in parallel to get the same power...again you have to add up the heater current the gain may be in heat spread over a larger area..
Here is an example
http://www.transcendentsound.com/Transcendent/Products.html
I have no connection to this link just for information..
Yes but using PL509 there are circuits on the Web..Heater is a lot less power..
I guess I should point out that when I say Use power...By running my OTL 6C33C on only one heater I save 100W of mains power..I think it sounds very good..I have the option of full power I only use it occasionally.. have spoken to others and they like /need the heat from the OTL because it helps heat the house in freezing conditions..Also to them an OTL running at 300-350W at idle is low power... I guess its probably similar to 845's etc although I haven't checked so don't flame me ...
Having said all this I love the sound...The heat is an added bonus I get to be nice an warm in the listening room..In winter I take the mobile central heating stat in with me so it turns off and freezes everyone else in the house...
In the UK we are using power savers at 11W to replace 100W standard lamps..solar power is the new way forward..
All just for fun....
Regards
M. Gregg
thanks forthat greg, yes we here in nz are into the energy saver lamps too, they have become more reliable than a few years ago and yes the power usage of such an amp would be high, but ill only use it cranked up... i promise hahahahaha thank you for the 509 tip!
Yes but using PL509 there are circuits on the Web..Heater is a lot less power..
I guess I should point out that when I say Use power...By running my OTL 6C33C on only one heater I save 100W of mains power..I think it sounds very good..I have the option of full power I only use it occasionally.. have spoken to others and they like /need the heat from the OTL because it helps heat the house in freezing conditions..Also to them an OTL running at 300-350W at idle is low power... I guess its probably similar to 845's etc although I haven't checked so don't flame me ...
Having said all this I love the sound...The heat is an added bonus I get to be nice an warm in the listening room..In winter I take the mobile central heating stat in with me so it turns off and freezes everyone else in the house...
In the UK we are using power savers at 11W to replace 100W standard lamps..solar power is the new way forward..
All just for fun....
Regards
M. Gregg
Not with PL500 or PL504 but with PL36 , look here if you want .
http://www.diyaudio.com/forums/tubes-valves/200823-my-new-circlotron-otl-tube-power-amp.html
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.