As far as what is loud, I don't have an SPL meter, so I judge it by other things.
Loud is when your wife comes upstairs from the other side of the house and closes the door to the room. It's when you feel the bass. It's when the dog barks at the speakers and runs out of the room.
Loud is when your wife comes upstairs from the other side of the house and closes the door to the room. It's when you feel the bass. It's when the dog barks at the speakers and runs out of the room.
Attachments
OK, so here's some quantitative measure of "how loud".
I downloaded the AudioTools app to my iPhone. Using an unweighted SPL setting, I got a 111 db peak while listening to Dire Strait's "The Man's Too Big." If you are familiar with the tune, you know where the peak occurred. (LOL).
The measurement was taken at the listening position, which is about 9 feet from the speakers.
Room temperature is 78 degrees. (It's cold here in McKinney today so my wife jacked up the heat.)
I am using Martin Logan Motion 12 speakers, which are rated at 92 db, 4 ohms.
If my math is correct 111 db corresponds to 64 watts from the amp, but remember the measurement was taken more than 1 meter from the speakers, so that's probably a bit low.
The most sensitive SPL meter I have is my Labrador Retriever, Maggie. (See avatar). She has been hiding under the bed since I took the measurement.
One other subjective observation: I listened to this with absolutely no discomfort. There was no strain at all. I've used this analogy before: it's like driving a 911 at 130 mph; it feels like 60 mph in a normal car. (I did it on the autobahn one time. It was worth the trip and the price of the rental. Crossed that one off my bucket list. We won't discuss how much beer I drank afterwards.)
Like I said, as smooth as Macallan 25. (Fifty bucks a shot, another bucket list item.)
One final note: based on some web research, listening at those levels causes damage in .9 minutes.
I am not sure if I am hard of hearing from listening to music or being married for 27 years.
You can't buy stuff this good; at least not on my budget.
Cheers.
I downloaded the AudioTools app to my iPhone. Using an unweighted SPL setting, I got a 111 db peak while listening to Dire Strait's "The Man's Too Big." If you are familiar with the tune, you know where the peak occurred. (LOL).
The measurement was taken at the listening position, which is about 9 feet from the speakers.
Room temperature is 78 degrees. (It's cold here in McKinney today so my wife jacked up the heat.)
I am using Martin Logan Motion 12 speakers, which are rated at 92 db, 4 ohms.
If my math is correct 111 db corresponds to 64 watts from the amp, but remember the measurement was taken more than 1 meter from the speakers, so that's probably a bit low.
The most sensitive SPL meter I have is my Labrador Retriever, Maggie. (See avatar). She has been hiding under the bed since I took the measurement.
One other subjective observation: I listened to this with absolutely no discomfort. There was no strain at all. I've used this analogy before: it's like driving a 911 at 130 mph; it feels like 60 mph in a normal car. (I did it on the autobahn one time. It was worth the trip and the price of the rental. Crossed that one off my bucket list. We won't discuss how much beer I drank afterwards.)
Like I said, as smooth as Macallan 25. (Fifty bucks a shot, another bucket list item.)
One final note: based on some web research, listening at those levels causes damage in .9 minutes.
I am not sure if I am hard of hearing from listening to music or being married for 27 years.
You can't buy stuff this good; at least not on my budget.
Cheers.
Thanks, I also have that app on Andriod phone and iPad. I'll do some tests here and report my results when I get my BA-3 back together (apart for cooling system mods)
You have been very helpful, but may I ask you go a little further and list any - if any - component and settings mods you made to get the output level you have? I'm aware I need to reduce the value of the resistors on the output boards, but still not clear on the other safe and effective variations.
You have been very helpful, but may I ask you go a little further and list any - if any - component and settings mods you made to get the output level you have? I'm aware I need to reduce the value of the resistors on the output boards, but still not clear on the other safe and effective variations.
All the gain comes from the front end, so that's where I will focus first.
I used stock values for all components on the front end board except P1 and P2. I went with 1K pots instead of 500 ohm, so I'd have more voltage range to bias Q3 and Q4.
I am running Q3 and Q4 at 100 ma. This gives 2.2V across R10 and R11. This does not affect gain, but it should reduce distortion compared to the "Stock" setting of 1V which gives about 50 ma bias.
The gain is roughly R13/R10, or 15. Since it's a push-pull stage you double that to get the full gain, which is 30X.
Make sure all your resistor values are correct. It's easy to mess up. For example if you accidently used a 220 ohm for R10 and R11 your bias current would be very low and the gain would only be 332/220. The other critical values are R5, R6, and R7. They should all be 100 ohms, which is about unity gain.
Now for the output stage.
Are you using the SE or push-pull output stage? I am using the push-pull, biased at 500 ma instead of 250ma. Once again, that will not affect gain, only distortion. 500 ma will also limit excursions into class AB, more so than 250 ma.
If you are using the SE output stage your output power will be lower than that of the push-pull stage. If you are going with the SE and driving a difficult speaker (like a 4 ohm nominal speaker) the amp will also distort sooner.
My voltage rails are about +/- 30V. If you are running a lower voltage your amp will clip sooner.
What impedance and sensitivity are your speakers? How big is the room? How far away is your listening position? All of those things impact SPL at a given location in a much bigger way than the amp itself.
I used stock values for all components on the front end board except P1 and P2. I went with 1K pots instead of 500 ohm, so I'd have more voltage range to bias Q3 and Q4.
I am running Q3 and Q4 at 100 ma. This gives 2.2V across R10 and R11. This does not affect gain, but it should reduce distortion compared to the "Stock" setting of 1V which gives about 50 ma bias.
The gain is roughly R13/R10, or 15. Since it's a push-pull stage you double that to get the full gain, which is 30X.
Make sure all your resistor values are correct. It's easy to mess up. For example if you accidently used a 220 ohm for R10 and R11 your bias current would be very low and the gain would only be 332/220. The other critical values are R5, R6, and R7. They should all be 100 ohms, which is about unity gain.
Now for the output stage.
Are you using the SE or push-pull output stage? I am using the push-pull, biased at 500 ma instead of 250ma. Once again, that will not affect gain, only distortion. 500 ma will also limit excursions into class AB, more so than 250 ma.
If you are using the SE output stage your output power will be lower than that of the push-pull stage. If you are going with the SE and driving a difficult speaker (like a 4 ohm nominal speaker) the amp will also distort sooner.
My voltage rails are about +/- 30V. If you are running a lower voltage your amp will clip sooner.
What impedance and sensitivity are your speakers? How big is the room? How far away is your listening position? All of those things impact SPL at a given location in a much bigger way than the amp itself.
Great, I'm about 40 minutes away from firing up the BA-3. I'll measure with the app and respond then. I do have a Radio Shack standard level meter but have to go get some fresh 9V batteries. I did do some reading in the last hour to see if I could determine the voltage at the input of the front end accurately, but don't think I have the necessary equipment - yet. New digital scope on the way.
Thanks
Thanks
Last edited:
BTW: I'm using Paul Carmody's Sunflowers (those are mine at the top of the page) but I have never been able to get him to provide the sensitivity spec. Maybe someone can decipher it from the article.
Last edited:
Wow! Big eye opener. For a standard I connected only a Sony DVP-NS400 through an Alps bare pot. The spec sheet does indeed say 2V (rms) @ 50 kilohoms with over 10 Kilohms Load impedance.
With the pot at max and reading at 9', the small home brew bench speakers read 79 db CWt/Avg. The Sunflowers read only 72 on the same CD. That's probably no where near what the Pass designs like.
My build is stock from the published (from the diyAudio store) BOM. The output is the SE version that I selected specifically for max heat output with the liquid cooling trials. It has a total of 24 devices on four boards.
The room is only 12' x 15' with 7' ceiling, but that may not be so important with the results discovered today. I'd like to avoid building another set of speakers (but will if necessary) if possible cause I like what I have when paired up with the MyRef Fremen Edition amps. They need attenuation only for loud. I don't think I could do this test (full pot) with those power amps without blowing or damaging the speakers.
I'm reminded of Nelsons comment " Driving 25 miles an hour with the gas pedal on the floor."
All suggestion welcomed
With the pot at max and reading at 9', the small home brew bench speakers read 79 db CWt/Avg. The Sunflowers read only 72 on the same CD. That's probably no where near what the Pass designs like.
My build is stock from the published (from the diyAudio store) BOM. The output is the SE version that I selected specifically for max heat output with the liquid cooling trials. It has a total of 24 devices on four boards.
The room is only 12' x 15' with 7' ceiling, but that may not be so important with the results discovered today. I'd like to avoid building another set of speakers (but will if necessary) if possible cause I like what I have when paired up with the MyRef Fremen Edition amps. They need attenuation only for loud. I don't think I could do this test (full pot) with those power amps without blowing or damaging the speakers.
I'm reminded of Nelsons comment " Driving 25 miles an hour with the gas pedal on the floor."
All suggestion welcomed
Attachments
Last edited:
OK, I'm a newbie to Nelson Pass amps. Been doing tube amps for about 15 years and I'm an EE by trade, but over the years I've come to know what I don't know. I'm getting real close to out of my depth for a 6-deep SE output stage, but I have a couple more questions.
What is the VA rating of your power transformer? Who made it? Do you know how much peak current it can handle?
How much total capacitance are you using?
What is your rail voltage?
What bias point are you using for the output FETs?
You may need some help from some of the people on the forum who are more experienced with your amp's configuration.
What is the VA rating of your power transformer? Who made it? Do you know how much peak current it can handle?
How much total capacitance are you using?
What is your rail voltage?
What bias point are you using for the output FETs?
You may need some help from some of the people on the forum who are more experienced with your amp's configuration.
Each side/channel has an Antek AN-5218
PS are 27,000 x 4 each side. They are producing ~ 21.2 VDC
The bias recommendation is what I think I'm seeking through these several posts.
PS are 27,000 x 4 each side. They are producing ~ 21.2 VDC
The bias recommendation is what I think I'm seeking through these several posts.
Attachments
If I have confused (bored or frustrated) anyone in my oft posted questions about output levels, let me try to clarify.
It's my take away from the gain structure article that keeping the gain of each segment in the signal path relatively high above the noise floor is a good thing. With that in mind -
As the front end of the BA-3 essentially operates as a pre, that's a good spot (or several spots) for tweaking.
Secondly, the resistors on the output boards can be adjusted to allow higher output.
The third option is a bump up from an external gain device before the front end.
And finally, more efficient speakers can/will increase the apparent SPL.
I hope I have that correct. I'm not trying to create a monster that can fill an auditorium, but feel the need for just about 10 % more SPL with a little headroom to spare. That would be my definition of loud in my room with my speakers.
I'm more than willing to experiment in all of the first three areas mentioned, but would prefer not to add an extra leg (pre w/gain) in the signal path if it can be avoided. What I read on the forum convinces me I already have what I need, but am still foggy on the exact method or best location to make the necessary adjustments. That may well be simply a personal preference, but I seek advice so I don't try to fix something that's not broken - and mess up this great sounding amp.
It's my take away from the gain structure article that keeping the gain of each segment in the signal path relatively high above the noise floor is a good thing. With that in mind -
As the front end of the BA-3 essentially operates as a pre, that's a good spot (or several spots) for tweaking.
Secondly, the resistors on the output boards can be adjusted to allow higher output.
The third option is a bump up from an external gain device before the front end.
And finally, more efficient speakers can/will increase the apparent SPL.
I hope I have that correct. I'm not trying to create a monster that can fill an auditorium, but feel the need for just about 10 % more SPL with a little headroom to spare. That would be my definition of loud in my room with my speakers.
I'm more than willing to experiment in all of the first three areas mentioned, but would prefer not to add an extra leg (pre w/gain) in the signal path if it can be avoided. What I read on the forum convinces me I already have what I need, but am still foggy on the exact method or best location to make the necessary adjustments. That may well be simply a personal preference, but I seek advice so I don't try to fix something that's not broken - and mess up this great sounding amp.
Last edited:
Hmm, in stereo, 92 dB + 92 dB = 10log[2]+92 = ~95 dB
If distance was ignored, i.e. @ 1 m, then 2^[[111 - 95]/3.01] = ~39.83 W
9 ft away though = 20log[~9/3.280833] = ~8.76 dB + 111 dB = ~119.76 dB/m
2^[[119.76 - 95]/3.01] = ~299.74 W
30^2/4 = 225 W, but room ambiance can easily make up the ~1.25 dB theoretical power shortage to keep the amp from clipping, though its dynamic headroom appears inadequate unless there’s at least +4 dB of room gain and/or the speakers are corner loaded.
GM
BA-3 Listening Session
I had a golden-eared non-technical friend of mine come give the BA-3 a listen last night.
He says he wants one, which is high praise coming from him.
I said my skills could be leveraged with beer and bratwurst.
Since he has an MBA, I told him he needs to ignore cost.
We will build it together.
Should be fun and a good opportunity to expose him to this arcane world we enjoy so much.
I had a golden-eared non-technical friend of mine come give the BA-3 a listen last night.
He says he wants one, which is high praise coming from him.
I said my skills could be leveraged with beer and bratwurst.
Since he has an MBA, I told him he needs to ignore cost.
We will build it together.
Should be fun and a good opportunity to expose him to this arcane world we enjoy so much.
