The 6.2V Zener is a better match for IRFP240 for biasing purposes.
9.1V Zener does not operate as intended in the modified circuit and the 5.1V Zener of the original circuit is slightly too small to get adequate Vgs for biasing
Last edited:
ask Full Range where I wrote about that .......
On this page of the F6 Ilustrated build guide - http://www.diyaudio.com/forums/pass-labs/277850-f6-illustrated-build-guide-3.html
Please note that ZM did not mention R13 - That was my decision to omit
FR
So are we hacking Papa's design here)), or is this a necessity due use of different transistors than in his original design?
You can stick with Papa's original design with 5.1V Zener if you feel that's what you want, but if you have a problem reaching 1.3A bias then you will already have a solution to refer to.
True. I have a 9.1V zener as per the BOM in the build guide. Will attempt to bias probably tomorrow night when I finish the amp.
When you say it does not operate as intended, is this just a matter of it being more difficult to get the bias right (and once right it doesn't really matter), or is this something that impacts the performance/sound?
When you say it does not operate as intended, is this just a matter of it being more difficult to get the bias right (and once right it doesn't really matter), or is this something that impacts the performance/sound?
The 9.1V Zener will not operate as intended with the resistor values shown in the circuit.
Just copy Zen Mod's circuit
Last edited:
So are we hacking Papa's design here
Let's just say that it's the same design but taking a slightly different road to the finish line
FR
Looks like I need another mouser order
Will I damage the amp by powering it up with the 9.1V zener?
Has someone let 6L6 know so he can modify his guide? I guess 95% of people building this amp are going by his guide and won't find this little nugget of info hidden deep inside the threads... According to 6L6 in the guide "everybody who has used 9.1V has had zero issues"
No it won't damage the amp. It will basically work off the trimpot as regular voltage divider. So it will bias up.Looks like I need another mouser order
Will I damage the amp by powering it up with the 9.1V zener?
Has someone let 6L6 know so he can modify his guide? I guess 95% of people building this amp are going by his guide and won't find this little nugget of info hidden deep inside the threads... According to 6L6 in the guide "everybody who has used 9.1V has had zero issues"
Make sure you have the trimpots in the correct position when turning on or you might cook a mosfet.
Yes, it will.
You are welcome to use a Zener value bigger than 5.1V and smaller than 9.1V, it's nowhere as critical as you are all making it out to be.
The zener and resistor (pot) must have enough voltage across it to make Vgs of the Mosfet and some more to get the bias set. The first few people who built it couldn't get enough bias with the stock zener value, so being Pass Amp builders, we reached into our boxes of bits and used 9.1V zener, as most Pass amps use them, and we all have extras.
Of course not.
Modify how? I'm unclear as to what you are asking.
At the time I wrote that with the collective experience of the various builders, it was true. I still stand by it.
Yes, it will.
You are welcome to use a Zener value bigger than 5.1V and smaller than 9.1V, it's nowhere as critical as you are all making it out to be.
I agree it's not that critical. In fact you can take the Zener out of the circuit all together and it will bias up with the trim pot.
My arguement is, and only to be pedantic about it is. I refuse to believe you will measure 9.1V across the 5k pot with 9.1V Zener inserted in series with 10k resistor.
My brain says you will measure between 7V and 8V. Either way it will still bias up.
Have you measured the voltage across the pot?
There is less than 1mA to feed the 9.1V Zener (rough estimate).
Last edited:
9V1 is perfectly viable , just having lesser part of trimpot available for fine tuning
anyway , as pico pointed out somewhere , when using 9V1 instead of 5V1 , series resistor must be decreased in value
with stock - 22Vdc - 10K - 5V1 5K trimpot , trimpot will take app. 1mA , leaving 0mA69 for zenner itself
when using 9V1 in place of 5V1 , thing will work , but somewhat poorly regulated , due to zenner starving for current
this is not nitpicking (I believe even not from resident PicoDodo) .... just think that Greedy Boyz need to learn few itsybitsys , along with Legoing their amps
it's Papa's approach , ne pas ?
or written in slightly different form
Pass
ne
pas
anyway , as pico pointed out somewhere , when using 9V1 instead of 5V1 , series resistor must be decreased in value
with stock - 22Vdc - 10K - 5V1 5K trimpot , trimpot will take app. 1mA , leaving 0mA69 for zenner itself
when using 9V1 in place of 5V1 , thing will work , but somewhat poorly regulated , due to zenner starving for current
this is not nitpicking (I believe even not from resident PicoDodo) .... just think that Greedy Boyz need to learn few itsybitsys , along with Legoing their amps
it's Papa's approach , ne pas ?
or written in slightly different form
Pass
ne
pas
no need for that
physics is physics , at least on elementary level
sorry - not elementary ..... basic
I too have used the 9.1 Volt Zener route and found the regulation was not that good and if you read the data sheet you will find they need a certain amount of current to be stable typically 10 to 30mA.
Seeing Zenmod's post of the resistor change to 3.3K ohms make sense.
Since I have used the HA5002 high speed buffer as the front end of my F6 I included a regulated +- 20 Volt supply and simply took a resistor divider to generate the bias voltage for the Crees.
Have not done listening tests yet to compare the two but will report back soon.
Seeing Zenmod's post of the resistor change to 3.3K ohms make sense.
Since I have used the HA5002 high speed buffer as the front end of my F6 I included a regulated +- 20 Volt supply and simply took a resistor divider to generate the bias voltage for the Crees.
Have not done listening tests yet to compare the two but will report back soon.
Ok to prove my point as simple as possible (and feel free to correct me, my brain runs on limited brain cells)
1) You have a 9.1V Zener in Parallel with 5k pot. So if Zener is operating correctly you must have 9.1V across 5k trimpot
2) I=V/R : Current through trimpot must be equal 9.1/5000 = 1.82mA
3) So to have 1.82mA at the pot it needs to travel through the 10k resistor. The potential drop across the 10k resistor will be V=IR, so V=0.00182x10000= 18,2V
4)And here is the straw that broke the camel's back. You have 23V rails. 23V - 18.2V = 4.8V. That means the Zener cannot even function as intended.
Which means what is happening in this part of the circuit is you have a 10k resistor + 5k pot in series (15k total). 5k is a third of 15k total resistance. So you have a voltage drop of 2/3 x 23V across the 10k resistor and 1/3 x 23V across the pot, in other words you have between 7V and 8V across the pot to bias the circuit and the 9.1V Zener has gone to sleep.
I'm quite sure this correct. There are other approaches to explain the same conundrum. I hope this explanation is the simplest (assuming I didn't stuff it up)
1) You have a 9.1V Zener in Parallel with 5k pot. So if Zener is operating correctly you must have 9.1V across 5k trimpot
2) I=V/R : Current through trimpot must be equal 9.1/5000 = 1.82mA
3) So to have 1.82mA at the pot it needs to travel through the 10k resistor. The potential drop across the 10k resistor will be V=IR, so V=0.00182x10000= 18,2V
4)And here is the straw that broke the camel's back. You have 23V rails. 23V - 18.2V = 4.8V. That means the Zener cannot even function as intended.
Which means what is happening in this part of the circuit is you have a 10k resistor + 5k pot in series (15k total). 5k is a third of 15k total resistance. So you have a voltage drop of 2/3 x 23V across the 10k resistor and 1/3 x 23V across the pot, in other words you have between 7V and 8V across the pot to bias the circuit and the 9.1V Zener has gone to sleep.
I'm quite sure this correct. There are other approaches to explain the same conundrum. I hope this explanation is the simplest (assuming I didn't stuff it up)
Last edited:
I'm pretty sure the only thing that will happen if you use 9.1V is the bias will have a significant amount of ripple since the zener wouldn't be doing anything (as designed the voltage will be about 7V across the diode so 5-7V zeners should work.) The amp won't catch on fire or anything but it is important to ensure the zener is effectively regulating the voltage. edit: pretty much like ZM et al have already mentioned
Last edited:
Continuing with 2pico's example above, I think the bias would be somewhat
sensitive to the voltage rails.
If your rail is V_0=23V, then you basically have 23/3 volts
across the 5K trimmer. You adjust the trimmer to set the bias. Let say you
end up with Vgs = 4.6 V. So the trimmer reduces the voltage to 0.6x the input
value. In other words, the Vgs ~ 0.6 * V_0/3.
Now, imagine if the rail voltage is not perfectly stable and it has gone up to 24V,
say caused by AC mains variation, then the Vgs ~ 0.6 * 24 /3 = 4.8V. I imagine
this is enough to cause a noticeable drift in bias.
Dennis
sensitive to the voltage rails.
If your rail is V_0=23V, then you basically have 23/3 volts
across the 5K trimmer. You adjust the trimmer to set the bias. Let say you
end up with Vgs = 4.6 V. So the trimmer reduces the voltage to 0.6x the input
value. In other words, the Vgs ~ 0.6 * V_0/3.
Now, imagine if the rail voltage is not perfectly stable and it has gone up to 24V,
say caused by AC mains variation, then the Vgs ~ 0.6 * 24 /3 = 4.8V. I imagine
this is enough to cause a noticeable drift in bias.
Dennis
Please post results. I also have that piece installed but will switch it out. Let us know what you find. Some have had it drift a bit and attributed it to this, I don't know. I don't think ZenMod would advise if it wasn't the right thing to do here. Similarly, if 9 volt item failed to work, 6L6 would not have said it would.
Russellc
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.