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 23 Jul 2019, 20:32

 All times are UTC - 6 hours [ DST ]

 Page 2 of 3 [ 28 posts ] Go to page Previous  1, 2, 3  Next
 Print view Previous topic | Next topic
Author Message
 Posted: 11 Jun 2019, 07:14

Joined: 19 Oct 2018, 15:30
Posts: 126
Location: Montréal, Québec
Suncalc wrote:
Gab wrote:
As you can see in the schematic description, both impedance and resistance are specified. I assume that since the transformer replaces the cathode resistor, it must have a precise resistance (dc).
Not quite.

The 6V6 cathode load is the series combination of R3 (1000Ω), R5 (355Ω), and the transformer primary dc resistance (250Ω) for a total of ≈1.6kΩ. This (along with the B+ voltage) sets the 6V6 quiescent plate voltage but NOT the grid bias. The grid bias is set by the resistive voltage divider formed by R3/R5/R6/R7. The 49.5mA cathode current drops 12.4 volts in the transformer primary which adds to the voltage set by the R6/R7 divider to set the cathode at 85v against the grid at 73v. Total grid bias is therefore -12.4v. (i.e. essentially -12.5v, the most common 6V6 grid bias.) Also note that the plate voltage works out to be 355-85=250v, the most common plate voltage for the 6V6 tube.

Thanks for sharing your knowledge. Could you just confirm what I understand from what you have said : since the primary resistance of the transformer isn’t the only thing setting the cathode load, a small difference won’t make the entire circuit go wrong.

Thanks again for your help ! It is very appreciated !

Top

 Posted: 11 Jun 2019, 07:15

Joined: 19 Oct 2018, 15:30
Posts: 126
Location: Montréal, Québec
This is a very uncommon and interesting circuit topology. I wonder why so little people have tried to build it !

Top

 Posted: 11 Jun 2019, 19:27
 Project Author

Joined: 06 Apr 2009, 10:08
Posts: 1586
Location: US Pacific Northwest
Gab wrote:
Could you just confirm what I understand from what you have said : since the primary resistance of the transformer isn’t the only thing setting the cathode load, a small difference won’t make the entire circuit go wrong.
Yes that's correct. A small difference really won't matter. But the bias is partially set based on the primary resistance so it will change things a little.
Gab wrote:
This is a very uncommon and interesting circuit topology. I wonder why so little people have tried to build it !
Using resistance ladders to set circuit voltages was a common method of design in the very early days of tubes. Here is an example of a two stage, direct coupled amplifier using this approach.
Attachment:
Screen Shot 2019-06-11 at 5.17.47 PM.png
The reasons for this were two fold. First, many early radios were run from batteries since electrical power was not ubiquitous. And since batteries were expensive, it just made sense to specify one with a high enough voltage for the B+ and find ways to use resistors to get the lower biasing voltages. Second, this method worked well with direct coupled amplifiers. In the early days of tubes, capacitors for coupling stages were expensive, and of rather poor quality. They were prone to leakage and early failure.

This was an interesting circuit in its day. However, given that we now have access to very high quality, inexpensive signal capacitors, this circuit is more of a curiosity than anything else. IMHO.

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

_________________
Matt

Top

 Posted: 11 Jun 2019, 19:56

Joined: 23 Feb 2017, 02:02
Posts: 500
Quote:
"Would the impedance of the output transformer go up if a resistor was added in series with it ? Since mine are 5k, they would get closer to the specified 6k."

No it will not go up, all your doign is reducing the power output into the speakers when you do that.
Further more you reduce the maximum voltage swing possible on the speakers.
Transformer impedance stays the same, but you just lose extra power on the 1k resistor to get 6k.

Should be able to do 5k instead of 6k, minor changes dosen't matter too much and if you know how to fine tune stuff thats even less of a plorbem. But you'll probally design your own amp anyways if you know to design.

The transformer resistance is specified in scheamtic, because that will effect the operating point in DC. All you do is just fine tune the resistors to controll grid cathode voltage to compensate if you have diffrent resistance.

Using Power formula I^2R to find power accros the resistor, dudes in the old days can't even get utmost basics right, R5 355R 10w should be only a 5w resitor.
(52*10^-3)^2*355 = 0.96W of power dispation in R5
(52*10^-3)^2*1000 = 2.7W of power dispation in R3

Top

 Posted: 12 Jun 2019, 06:44

Joined: 19 Oct 2018, 15:30
Posts: 126
Location: Montréal, Québec
ILoveHiFi wrote:
Quote:
"Would the impedance of the output transformer go up if a resistor was added in series with it ? Since mine are 5k, they would get closer to the specified 6k."

No it will not go up, all your doign is reducing the power output into the speakers when you do that.
Further more you reduce the maximum voltage swing possible on the speakers.
Transformer impedance stays the same, but you just lose extra power on the 1k resistor to get 6k.

Should be able to do 5k instead of 6k, minor changes dosen't matter too much and if you know how to fine tune stuff thats even less of a plorbem. But you'll probally design your own amp anyways if you know to design.

The transformer resistance is specified in scheamtic, because that will effect the operating point in DC. All you do is just fine tune the resistors to controll grid cathode voltage to compensate if you have diffrent resistance.

Using Power formula I^2R to find power accros the resistor, dudes in the old days can't even get utmost basics right, R5 355R 10w should be only a 5w resitor.
(52*10^-3)^2*355 = 0.96W of power dispation in R5
(52*10^-3)^2*1000 = 2.7W of power dispation in R3

I clearly don’t know how to design complete circuits yet. I wish I could though ! I need a lot more reading and practicing before I’m comfortable enough with vaccum tubes principles. But, i’m getting closer to this goal each time you guys share your knowledge with me on the forum ! I really appreciate it !

Top

 Posted: 12 Jun 2019, 07:13

Joined: 19 Oct 2018, 15:30
Posts: 126
Location: Montréal, Québec
Suncalc wrote:
Gab wrote:
Could you just confirm what I understand from what you have said : since the primary resistance of the transformer isn’t the only thing setting the cathode load, a small difference won’t make the entire circuit go wrong.
Yes that's correct. A small difference really won't matter. But the bias is partially set based on the primary resistance so it will change things a little.
Gab wrote:
This is a very uncommon and interesting circuit topology. I wonder why so little people have tried to build it !
Using resistance ladders to set circuit voltages was a common method of design in the very early days of tubes. Here is an example of a two stage, direct coupled amplifier using this approach.
Attachment:
Screen Shot 2019-06-11 at 5.17.47 PM.png
The reasons for this were two fold. First, many early radios were run from batteries since electrical power was not ubiquitous. And since batteries were expensive, it just made sense to specify one with a high enough voltage for the B+ and find ways to use resistors to get the lower biasing voltages. Second, this method worked well with direct coupled amplifiers. In the early days of tubes, capacitors for coupling stages were expensive, and of rather poor quality. They were prone to leakage and early failure.

This was an interesting circuit in its day. However, given that we now have access to very high quality, inexpensive signal capacitors, this circuit is more of a curiosity than anything else. IMHO.

To me, the interesting parts of the circuit are the use of direct coupling and the cathode follower output stage.

My main system is direct coupled and I really like the sound of it. The power amp is an Adcom GFA-555 with an Adcom GTP-500II preamp, and both are direct coupled (if you use the "lab" output of the GTP). I believe that there is no such thing as a perfect capacitor, or any electrical component for that matter. It may only have a negligible impact, but I’m a firm believer of the less is more philosophy. I don’t have all the knowledge to back this statement, I just see this as something logical.

The other interesting part, the cathode follower output stage, leaves me puzzled. I thought that a cathode follower didn’t provide any voltage gain, but this amp has an output of 4.5w according to the author. Also, it is not clear to me if the 6V6 is used as a pentode or a triode.

Finally, your comment confirm what I had on my mind : looks like an old fashion way to get this amp working. Power supply parts are very minimalistic, one more time because of the availability, quality and price of capacitors in that time. I’d like to build the amp, but taking advantage of the availability of modern parts. The circuit would become more of an inspiration than anything else, but i’d like to use something like an ef86 with el84 output stage, and a power supply with multiple voltages output replacing the resistor ladder. The power transformer I plan on using have multiple secondaries, so it could make sense To build the supply that way. On the other hand, it would only remove the resistor ladder to add many parts in the supply, so that is not simpler in any way.

Let me know what you think !

Top

 Posted: 12 Jun 2019, 18:52

Joined: 23 Feb 2017, 02:02
Posts: 500
Before you belive no perfect capacitor exist, you need to try erse pulse x. (1-2usd each)
Even my latest pre amps with loads of NFB more than 200dB to make distoriton like fully zero.
I use erse pulse x for the input and output caps.

These have just very little tone to the music, but its a plesant postive tone like valves adding postive sound.
However overtime with burn in it will just sound like a capcitor that is perfect with zero tone.
Also sounds like a capacitor that is perfect, zero distortion.

Top

 Posted: 12 Jun 2019, 19:22

Joined: 23 Feb 2017, 02:02
Posts: 500
If you want to use a totally diffrent output valve thats a big job, curent one has the transformer and cathode resistors matched so it can get maximum power out of the circuit.
Doing this is not easy.

Top

 Posted: 13 Jun 2019, 06:54

Joined: 19 Oct 2018, 15:30
Posts: 126
Location: Montréal, Québec
I’ll try the circuit as it is first. I’m trying to design a power supply for it with PSUD2. It is pretty straightforward, i’m just not shure how much ripple would be acceptable for that circuit, considering a 355V B+. Could you give me an approximation in % or in mV ? Thanks !

Edit : I’ve read on a guitar amp site that 10% is acceptable for a push pull amp, and 5% for a single ended circuit. It seemed rather high so I tried to aim for 0.05%, wich gave me a max ripple voltage of 177mV. I was able to bring the ripple down to below 20mV with one LC filter. The supply stabilizes in less than 1mS. Would this be acceptable ?

Top

 Posted: 13 Jun 2019, 20:24

Joined: 23 Feb 2017, 02:02
Posts: 500
For supply ripple, all it means is noise on the output when you have no signal, some buzz or hiss.
For testing don't need to have super low noise supply.
Test and build have sound and the one works then you go big and fancy on the supply

https://i.stack.imgur.com/IfB18.jpg
https://encrypted-tbn0.gstatic.com/imag ... XMx1QH4Ur- (normally use this one to calc ripple)
To calculatre ripple without any rc filter, theres two formulas that should give you same results

f is mains frequency 60 or 50hz
Don't get units wrong, for 100uF this = 100*10^-6 F

Top

 Display posts from previous: All posts1 day7 days2 weeks1 month3 months6 months1 year Sort by AuthorPost timeSubject AscendingDescending
 Page 2 of 3 [ 28 posts ] Go to page Previous  1, 2, 3  Next

 All times are UTC - 6 hours [ DST ]