NEW  Matt presents bias and operation data for the 6V6 tube in SE operation - 6V6 Single-Ended (SE) Ultra Linear (UL) Bias Optimization.

# DIY Audio Projects Forum

Welcome to the DIY Audio Projects Message Forum. Use these forums to discuss Hi-Fi audio and to share your DIY Audio Projects. Registration is free and required to post messages and view the file attachments. Registration will only take a minute and registered users do not see any advertisements. After you have completed the online registration process, check your email (including spam/junk folder) for the verification email to activate your account. New members are under moderation - so your posts will not be visible until approved by a moderator. See the Read Me 1st, Forum RULES and Forum FAQ to get started on the forum.

 It is currently 10 Aug 2020, 00:56

 All times are UTC - 6 hours [ DST ]

 Page 1 of 2 [ 14 posts ] Go to page 1, 2  Next
 Print view Previous topic | Next topic
Author Message
 Post subject: Coupling capacitorPosted: 21 Feb 2015, 07:44

Joined: 04 Feb 2013, 06:52
Posts: 37
Location: North Wales
How does one determine the size of a coupling capacitor between the stages of an amplifier?
I understand the dielectric and quality have a large bearing on the sound of the amp. That the voltage rating needs to be above the actual voltage at that point in the circuit.
I have a schematic which shows 0.1 to 1 micro farad (400V) implying anything between will work, but what will produce the best sound?

Top

 Post subject: Re: Coupling capacitorPosted: 21 Feb 2015, 10:49

Joined: 27 Jun 2011, 10:13
Posts: 146
Location: PA
There is a lot of math that goes with this so I will keep it simple. One the bigger the cap the better the low end response. That is lower resistance to lower freq. That being said do not use the same value coupling cap through out the amp as this can lead to R/C NETWORK oscillator. What is usually done is small cat in the front end then bigger in later stages. What I use are .22.then .25 then .33. But that is me. I have seen from .1 in front end to .22 in the driver stage. A good starting point is the RCA Radiotron designer handbook that can explain it a lot more. I just tried to keep it simple.

Top

 Post subject: Re: Coupling capacitorPosted: 21 Feb 2015, 13:02
 Project Author

Joined: 06 Apr 2009, 10:08
Posts: 1677
Location: US Pacific Northwest
wolfdog428 wrote:
There is a lot of math that goes with this...
Well, not really. And don't worry about the "R/C network oscillation". Oscillation in the amp has virtually nothing to do with the relative values of the coupling caps. These values are set based on the components around them, not the coupling of the last or next stage.

And there are two conditions to consider. The first is low end rolloff. You want a low rolloff frequency to support good bass response. Most people generally take this to mean only fractions of a dB loss at 20Hz. Chose too small a capacitor and the low end suffers. The second thing to consider is bias excursion and blocking. This happens during mild to moderate overdrive conditions and causes loss of amplification for short periods of time. Chose a capacitor too large and this becomes an issue robbing the amp of good transient response. Luckily, both conditions are easy to avoid.

Lets start with low end rolloff frequency. The low end roll off is set by the output impedance of the driving stage, the input impedance of the stage being driven (usually just the grid resistor), and the value of the coupling cap. Consider the case of two simple common cathode triode stages tied together.
Attachment:
f-lo due to Cc.jpg
The frequency given by this relation is the -3dB rolloff point. So if you want a small loss at 20Hz you might want this frequency to be between 5Hz and 10Hz.

Consider a simple example. Lets say that both stages are 12AU7s with moderate bias levels. RL1 is 50K and, at this bias level, rp of the driving stage is 10.0kΩ. This makes the driving stage output impedance 8.33kΩ (50kΩ||10kΩ). And lets say that the second stage grid resistor is 330kΩ. If I choose a 0.1µf coupling capacitor, then the -3dB rolloff frequency is 4.7Hz.

Now lets consider the other factor, bias excursions. Bias excursions happen when a large signal drives an input of a capacitively coupled tube stage into positive bias and the grid starts to draw current. When this happens, the DC voltage on the coupling capacitor which is normally fixed, begins to rise due to the charge transfer. This drives the bias on the grid down and forces the tube closer to cutoff. In extreme cases, the bias excursion drives the tube entirely to cutoff and the amp stops outputting signal.

So there are two ways to control the excursion, one is to make sure that the recovery time is short so that when excursions happen they clear quickly, the other is to slow the rate of excursion in the first place, so that excursions (when they do happen) are smaller that they would otherwise be. So, our design process should strive to lengthen bias excursion times and minimize bias recovery times, all while meeting the bandwidth requirements.

Now, bias excursion times are always shorter (or equal to) bias recovery times. The bias recovery time constant, τr, is Cc (the coupling capacitor) times the sum of Ro (the output impedance of the driving stage) and Rg (the grid leak resistor on the stage in question). But the excursion time constant, τe, is Cc (the coupling capacitor) times the sum of Ro (the output impedance of the driving stage) and the parallel combination of Rg and Rgs (the grid stopper resistor). Mathematically...

τr = Cc (Ro + Rg)
τe = Cc (Ro + Rg||Rgs)

So the design process is to chose Cc and Rg to make the bias recovery time as short as possible (personally I keep it below about 30mS) while still meeting the low frequency roll off requirements and then use a grid stopper to make the excursion time constant as long as reasonably possible, while not compromising the upper frequency roll off, so that it requires longer periods of overdrive to drive up the voltage on the coupling capacitor.

I like to use what I call the 400 beats per minute test. (400 bpm is about the fastest tempo you’ll commonly find in most music.) If a bias excursion occurs, I don’t want to “miss a beat” as it were. This means that I want the bias excursion recovery time to not exceed one beat at this fast tempo. This places the five time constant limit at 150mS.

So lets consider our example from above, using the values stated above, τr becomes 33.8mS for a total recovery time of 169mS. This is a little longer than I prefer, but not bad overall.

The other relation you need to know is that bias excursion time and low frequency are intimately linked. τr = 1/(2π * f-lo). This means that if you want to meet the limiter bias excursion time of 30mS then your rolloff frequency should be no lower than 5.3Hz. This frequency is still low enough to support very good low end performance yet high enough to help ensure short bias recovery time and hence good transient performance.

Does this make sense?

This post has a file attachment. Please login or register to access it. Only Registered Members may view attached files.

_________________
Matt

Top

 Post subject: Re: Coupling capacitorPosted: 21 Feb 2015, 13:13

Joined: 08 Aug 2009, 03:11
Posts: 2229
Location: Chilliwack, BC
It's good to overrate this capacitor, voltage-wise as well. Choose a voltage where under the worst fault conditions, the rating won't be exceeded. For tube amps, this would be a calculated max B+ with no load against maximum V- bias with no load.

Cheers!

_________________
-= Gregg =-
* Ratings are for transistors - tubes have guidelines*
Home: GeeK ZonE
Work: Classic Valve Design

Top

 Post subject: Re: Coupling capacitorPosted: 21 Feb 2015, 13:21
 Project Author

Joined: 06 Apr 2009, 10:08
Posts: 1677
Location: US Pacific Northwest
Geek wrote:
t's good to overrate this capacitor, voltage-wise as well. Choose a voltage where under the worst fault conditions, the rating won't be exceeded. For tube amps, this would be a calculated max B+ with no load against maximum V- bias with no load.
This is an EXCELLENT point!

Remember that before the tubes heat up you may have minimal or no plate current flow. This means that the driver side of the coupling cap is at the B+ voltage, not the steady state plate voltage. Add to this that the B+ voltage may be higher than steady state because the power stage is not drawing current. Together this means that at startup, the coupling cap may actually be seeing DC voltages GREATER than the steady state B+ voltage.

Great catch Gregg!

_________________
Matt

Top

 Post subject: Re: Coupling capacitorPosted: 21 Feb 2015, 17:36

Joined: 04 Feb 2013, 06:52
Posts: 37
Location: North Wales
Yes thanks, once I logged in and could see the diagram, the maths became clear.

Top

 Post subject: Re: Coupling capacitorPosted: 14 Mar 2015, 14:46

Joined: 12 Mar 2015, 17:24
Posts: 58
Location: Ennis MT
A good point in case here.
I've been working on an amp circuit for a mono block, a parallel push pull amp using 4x KT120's, giving about 125W. The original circuit used two BIAS adjustment trimmer for each pair of output tubes. I wanted individual BIAS adjustment, so changed the BIAS circuit accordingly.
Now, the original circuit called for two 0.33uF coupling caps. By going to individual BIAS adjustment, I had to add another two coupling caps to the circuit, plus an extra set of trimmer etc.. Somewhere I screwed up with my figures and retained the 0.33uF values. Well, I ended up having some pretty serious blocking distortion! After discussions with friends and reading up on blocking distortion and its causes, I changed the value of the coupling caps to 0.1uF as well as lowering the BIAS adjustment resistors, problem solved!
It ended up not changing the overall sound of the amp too much either.

Top

 Post subject: Re: Coupling capacitorPosted: 15 Mar 2015, 09:10

Joined: 27 Jun 2011, 10:13
Posts: 146
Location: PA
For coupling caps I use only Pio or new Jupiter wax caps. I do not go higher in voltage that is needed. But I do use larger value caps. For one reason. More low end! I do not use any sub woofer. I do not think they have any place in HiFi. Just as a mp3 or a phone for listing to is. The reason is I use big Cerwin Vega speakers. As do most who buy my amps. Like my they are old school and do not just want to here bass but want to feel it. I ran out of .22 caps on my last builds and put .1 in the first stage and what I got was a amp with 20 watts less output at full power in the lower end while the mids and highs were un changed. So I had to order some .22's and replace the .1's and then the low end came back to 100 watts at full power. My amps are for one thing and that is music. Not that cram that sounds like two monkeys in a garbage can or some thing that if woud play it back words you get your job back your dog does not die and you do not wreck your truck. The people who buy my amps listen to any thing from Abba to ZZ top and Led Zepplin and Betoven and Wagner. I have been doing this for over 40 years. If you want to limit you amp to 400 hertz that's in the low end up to you.

Top

 Post subject: Re: Coupling capacitorPosted: 15 Mar 2015, 10:00

Joined: 12 Mar 2015, 17:24
Posts: 58
Location: Ennis MT
wolfdog428 wrote:
For coupling caps I use only Pio or new Jupiter wax caps. I do not go higher in voltage that is needed. But I do use larger value caps. For one reason. More low end! I do not use any sub woofer. I do not think they have any place in HiFi. Just as a mp3 or a phone for listing to is. The reason is I use big Cerwin Vega speakers. As do most who buy my amps. Like my they are old school and do not just want to here bass but want to feel it. I ran out of .22 caps on my last builds and put .1 in the first stage and what I got was a amp with 20 watts less output at full power in the lower end while the mids and highs were un changed. So I had to order some .22's and replace the .1's and then the low end came back to 100 watts at full power. My amps are for one thing and that is music. Not that cram that sounds like two monkeys in a garbage can or some thing that if woud play it back words you get your job back your dog does not die and you do not wreck your truck. The people who buy my amps listen to any thing from Abba to ZZ top and Led Zepplin and Betoven and Wagner. I have been doing this for over 40 years. If you want to limit you amp to 400 hertz that's in the low end up to you.

that makes perfect sense, however, in my case it was not a matter of more or less bass etc., it was a matter of having to change the values to make it work. And as I said in the end, it made very little difference in the overall sound re production.
I too only use Russian PIO's for all my coupling caps, I find they are the best sounding, (to me), since sound is so extremely subjective!

Top

 Post subject: Re: Coupling capacitorPosted: 07 Apr 2015, 11:07

Joined: 15 Nov 2013, 10:16
Posts: 250
wolfdog428 wrote:
For coupling caps I use only Pio or new Jupiter wax caps. I do not go higher in voltage that is needed. But I do use larger value caps. For one reason. More low end! I do not use any sub woofer. I do not think they have any place in HiFi. Just as a mp3 or a phone for listing to is. The reason is I use big Cerwin Vega speakers. As do most who buy my amps. Like my they are old school and do not just want to here bass but want to feel it. I ran out of .22 caps on my last builds and put .1 in the first stage and what I got was a amp with 20 watts less output at full power in the lower end while the mids and highs were un changed. So I had to order some .22's and replace the .1's and then the low end came back to 100 watts at full power. My amps are for one thing and that is music. Not that cram that sounds like two monkeys in a garbage can or some thing that if woud play it back words you get your job back your dog does not die and you do not wreck your truck. The people who buy my amps listen to any thing from Abba to ZZ top and Led Zepplin and Betoven and Wagner. I have been doing this for over 40 years. If you want to limit you amp to 400 hertz that's in the low end up to you.

I dunno, coupling capacitor selection is pretty subjective isn't it?.. Your LF rolloff is set based on the RC network not just that coupling cap.. So if your grid leak resistor in the next stage is really high, then you don't need as large a coupling capacitor at that point in the amp.. If it's a really small grid leak resistor in the next stage then maybe you do need a rather large value..

To simply replace 0.1uF caps with 0.22uF without taking anything else into consideration doesn't really seem like the best route to go in my experience..

Top

 Display posts from previous: All posts1 day7 days2 weeks1 month3 months6 months1 year Sort by AuthorPost timeSubject AscendingDescending
 Page 1 of 2 [ 14 posts ] Go to page 1, 2  Next

 All times are UTC - 6 hours [ DST ]

#### Who is online

Users browsing this forum: No registered users and 13 guests

 You cannot post new topics in this forumYou cannot reply to topics in this forumYou cannot edit your posts in this forumYou cannot delete your posts in this forumYou cannot post attachments in this forum

 Jump to:  Select a forum ------------------ DIY Forums    Vacuum Tubes / Thermionic Valves    Solid State / Op-Amps / Chip Amps / Class-D    Speakers, Subwoofers and Headphones    Audio Sources and Signal Processing    Audio Cables and Power Cords    DIY Radio    DIY Electronics and Components    Chassis, Wood/Metalworking and Tools Hi-Fi and Audiophile Forums    Hi-Fi Equipment and Audiophile Talk    Audiophile Music and Audio Rooms    For Sale / Wanted / Group Buys    Off-topic and Introductions    Site Updates, News and Announcements