EL84 Amp - Baby Huey - Page 32

gingertube · 4th November 2007, 09:15 PM

Adam,
Just leave the 1K + 470 Ohm going to 0V and simply disconnect the 12K.

I try to run no or low global feedback always. I only put it in if the speakers sound underdamped - that is woolly or boomy bass and/or exagerated top end.

I find that anymore than about 4 to 5 dB of global feedback seems to shrink the stereo image.

Increasing the cross connect shunt feedback set resistor cleans up the output tubes at the expense of putting more demands on the diff amp / splitter. I tried 47K which was too much - sounded very "dry" and jusy like a solid state amp. 30 K was what I felt was the best.

For 8 Ohm Output just connect to 8 Ohm tap. My output trannies did'nt have an 8 Ohm tap - just 2 windings which go in parallel for 4 Ohm or in series for 16 Ohms.

Cheers,
Ian

LowRedMoon · 4th November 2007, 11:38 PM

Thanks for the reply, Ian.

for kicks, I'm going to try it both with and without global feedback.

If I'm running it into 8 ohm speakers, should the resistor remain at 12K?

I have a JJ 100uF+100uF 500V can. Is there any harm in using one of these for the 100uF caps instead of 2 separate units?

One more thing: Are all resistors whose watt ratings are not marked intended to be 1/2 watters? I could run the math for them all, but I'm sure you already have. I'm going to check the cathode R's just to be sure.

Building this one starts tonight. I'll post my results.

Thanks again,

Adam

gingertube · 5th November 2007, 12:07 AM

Adam,

yes any resistor without a power rating marked on the circuit can be 1/2 watt.

The JJ 100uF + 100uF electrolytics should be fine.

I assume you are proposing to take the global feedback from an 8 Ohm output and want to know what the new feedback resistor should be. What you need to achieve is to have the same feedback voltage at the grid.

Math:
Existing situation - 4 Ohm Output
Assume Output Power is approx 10 Watts (full power)
Power = V squared / R
rearrange
=> V = square root (P x R)
10 watts into 4 Ohm means 6.32 Volts RMS
Then by voltage divider, feedback voltage at the grid is
6.32 [470 / (12K + 470)] = 0.238 Volts RMS

New situation - 8 Ohm Output
10 watt into 8 Ohms means 8.94 Volts RMS
Then you want 8.94[470 / (Rfb + 470)] = 0.238V
to give the same feedback voltage at the grid.
Solving for Rfb (the new feedback resistor)
X multiply to get
8.94 x 470 = 0.238 (Rfb + 470)
=> 4201.8 = 111.9 + 0.238Rfb
=> 4089.9 = 0.238Rfb
=> Rfb = 17.1K

Use either 16K or 18K in place of the 12K.

Hope this makes sense to you.

Cheers,
Ian

LowRedMoon · 5th November 2007, 12:15 AM

Thank you for taking time to post the math. I've added it to my notes.

Quote:

Add a 630V rated film cap across each 100uF section...

Would this be in omission of the 0.22uF/400V that is already marked on the schematic?

gingertube · 5th November 2007, 12:33 AM

Too fast for me - I went back to the schematic and noted the caps you mentioned so edited my post to delete that comment. You got there before my edit.
The 0.22uF/400V will be fine.

To keep you up to date:
I'm half way through a build of a couple of new Baby Huey Mono Blocks where I'm throwing everything I can think of into the mix.

Basically I'm trying a choke input power supply with bigger electrolytics (Hammond 8H 150mA choke to JJ 470uF 500V Electrolytic). I'm changing the EL84s to fixed bias and have added ZVN0545A Mosfet Source Followers in each side to drive the output tubes. Once they are done I'll let you know if the added complexity/cost has provided any added performance. The original Baby Huey remains intact (in fact I've been running it on my system for the last week or 2) so I will be able to do some simple A/B listening tests. Circuit for these monoblocks looks similar to the 6V6 variant I posted on page 3 but with EL84 and some NOS 12AX7A.

Cheers,
Ian

LowRedMoon · 5th November 2007, 01:23 PM

Ian,

I've re-read through most of your previous posts and have a question about your feedback tweaking.

When making adjustments to the shunt feedback, were you operating with global feedback in place as well?

I intend to start with global in place for starters. The original schematic showed a shunt resistor vale of 16K. A cursory browse through my parts box shows that I have plenty of fine Mills NIWW's in the 22K-30K range, but nothing around 16K except 2 18K's which I intend to use in the global loop.

Is it OK to tweak the shunt resistor WITH global feedback or does this complicate things?

For the record - I'm going for a smooth, tonally lifelike, easy-to-listen-to sort of sound with as much articulation as possible without sounding "cold." Pretty typical requirements, I'm sure.

Gratefully,

Adam

gingertube · 5th November 2007, 09:45 PM

Adam,
No - connecting global fedback if required is always the last thing you do. Tweak the shunt feedback first and then if you think you still need it add the global feedback.

If you end up adding global feedback you can go back and re-tweak the shunt feedback later. Up to about 4.5 dB of global feedback does not change the imaging too much and that amount of GNFB will reduce the output impedance to 0.6 of the open loop level. Thats often worth having.

Cheers,
Ian

ctaudio · 5th November 2007, 11:17 PM

Quote:

Originally posted by gingertube
Basically I'm trying a choke input power supply with bigger electrolytics (Hammond 8H 150mA choke to JJ 470uF 500V Electrolytic). I'm changing the EL84s to fixed bias and have added ZVN0545A Mosfet Source Followers in each side to drive the output tubes.

I can't wait to see your new design. I am now on my fourth build of your Baby Huey. This time I'm putting it into a real chassis with some large computer grade power supply caps (5500uF). Those caps make a very noticeable difference in the bass.

Thanks again for a great project.

LowRedMoon · 5th November 2007, 11:39 PM

Here I go again - soaking this thread with simple questions...

The optional cap in the CCS does not have its polarity marked. I'm using a Nicholson Muse electrolytic. Which direction should I orient this one?

I too am looking forward to further improvements on this amp. Now if I can just get the current iteration in operation...

Ha det bra!

Adam

gingertube · 6th November 2007, 02:41 AM

Adam,
No such thing as a simple question - just simple answers.

Cap is wired +ve to the base of the bottom transistor. What is not marked on the schematic is that the voltage doubled supply from the 5V winding is generating a negative power supply to the bottom of that CCS circuit. Can't recall the exact voltage (and anyway its not that critical) but its about -12Volts.

Cheers,
Ian

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Red Light District/Baby Huey (EL84 P-P) Output Iron	dsavitsk	Tubes / Valves	3	14th February 2009 01:21 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

4th November 2007, 09:15 PM	#311
gingertube diyAudio Member Join Date: Oct 2004 Location: Adelaide South Oz	Adam, Just leave the 1K + 470 Ohm going to 0V and simply disconnect the 12K. I try to run no or low global feedback always. I only put it in if the speakers sound underdamped - that is woolly or boomy bass and/or exagerated top end. I find that anymore than about 4 to 5 dB of global feedback seems to shrink the stereo image. Increasing the cross connect shunt feedback set resistor cleans up the output tubes at the expense of putting more demands on the diff amp / splitter. I tried 47K which was too much - sounded very "dry" and jusy like a solid state amp. 30 K was what I felt was the best. For 8 Ohm Output just connect to 8 Ohm tap. My output trannies did'nt have an 8 Ohm tap - just 2 windings which go in parallel for 4 Ohm or in series for 16 Ohms. Cheers, Ian

4th November 2007, 11:38 PM	#312
LowRedMoon diyAudio Member Join Date: Jan 2006 Location: Upstate New York	Thanks for the reply, Ian. for kicks, I'm going to try it both with and without global feedback. If I'm running it into 8 ohm speakers, should the resistor remain at 12K? I have a JJ 100uF+100uF 500V can. Is there any harm in using one of these for the 100uF caps instead of 2 separate units? One more thing: Are all resistors whose watt ratings are not marked intended to be 1/2 watters? I could run the math for them all, but I'm sure you already have. I'm going to check the cathode R's just to be sure. Building this one starts tonight. I'll post my results. Thanks again, Adam

5th November 2007, 12:07 AM	#313
gingertube diyAudio Member Join Date: Oct 2004 Location: Adelaide South Oz	Adam, yes any resistor without a power rating marked on the circuit can be 1/2 watt. The JJ 100uF + 100uF electrolytics should be fine. I assume you are proposing to take the global feedback from an 8 Ohm output and want to know what the new feedback resistor should be. What you need to achieve is to have the same feedback voltage at the grid. Math: Existing situation - 4 Ohm Output Assume Output Power is approx 10 Watts (full power) Power = V squared / R rearrange => V = square root (P x R) 10 watts into 4 Ohm means 6.32 Volts RMS Then by voltage divider, feedback voltage at the grid is 6.32 [470 / (12K + 470)] = 0.238 Volts RMS New situation - 8 Ohm Output 10 watt into 8 Ohms means 8.94 Volts RMS Then you want 8.94[470 / (Rfb + 470)] = 0.238V to give the same feedback voltage at the grid. Solving for Rfb (the new feedback resistor) X multiply to get 8.94 x 470 = 0.238 (Rfb + 470) => 4201.8 = 111.9 + 0.238Rfb => 4089.9 = 0.238Rfb => Rfb = 17.1K Use either 16K or 18K in place of the 12K. Hope this makes sense to you. Cheers, Ian

5th November 2007, 12:33 AM	#315
gingertube diyAudio Member Join Date: Oct 2004 Location: Adelaide South Oz	Too fast for me - I went back to the schematic and noted the caps you mentioned so edited my post to delete that comment. You got there before my edit. The 0.22uF/400V will be fine. To keep you up to date: I'm half way through a build of a couple of new Baby Huey Mono Blocks where I'm throwing everything I can think of into the mix. Basically I'm trying a choke input power supply with bigger electrolytics (Hammond 8H 150mA choke to JJ 470uF 500V Electrolytic). I'm changing the EL84s to fixed bias and have added ZVN0545A Mosfet Source Followers in each side to drive the output tubes. Once they are done I'll let you know if the added complexity/cost has provided any added performance. The original Baby Huey remains intact (in fact I've been running it on my system for the last week or 2) so I will be able to do some simple A/B listening tests. Circuit for these monoblocks looks similar to the 6V6 variant I posted on page 3 but with EL84 and some NOS 12AX7A. Cheers, Ian

5th November 2007, 01:23 PM	#316
LowRedMoon diyAudio Member Join Date: Jan 2006 Location: Upstate New York	Ian, I've re-read through most of your previous posts and have a question about your feedback tweaking. When making adjustments to the shunt feedback, were you operating with global feedback in place as well? I intend to start with global in place for starters. The original schematic showed a shunt resistor vale of 16K. A cursory browse through my parts box shows that I have plenty of fine Mills NIWW's in the 22K-30K range, but nothing around 16K except 2 18K's which I intend to use in the global loop. Is it OK to tweak the shunt resistor WITH global feedback or does this complicate things? For the record - I'm going for a smooth, tonally lifelike, easy-to-listen-to sort of sound with as much articulation as possible without sounding "cold." Pretty typical requirements, I'm sure. Gratefully, Adam

5th November 2007, 09:45 PM	#317
gingertube diyAudio Member Join Date: Oct 2004 Location: Adelaide South Oz	Adam, No - connecting global fedback if required is always the last thing you do. Tweak the shunt feedback first and then if you think you still need it add the global feedback. If you end up adding global feedback you can go back and re-tweak the shunt feedback later. Up to about 4.5 dB of global feedback does not change the imaging too much and that amount of GNFB will reduce the output impedance to 0.6 of the open loop level. Thats often worth having. Cheers, Ian

5th November 2007, 11:39 PM	#319
LowRedMoon diyAudio Member Join Date: Jan 2006 Location: Upstate New York	Here I go again - soaking this thread with simple questions... The optional cap in the CCS does not have its polarity marked. I'm using a Nicholson Muse electrolytic. Which direction should I orient this one? I too am looking forward to further improvements on this amp. Now if I can just get the current iteration in operation... Ha det bra! Adam

6th November 2007, 02:41 AM	#320
gingertube diyAudio Member Join Date: Oct 2004 Location: Adelaide South Oz	Adam, No such thing as a simple question - just simple answers. Cap is wired +ve to the base of the bottom transistor. What is not marked on the schematic is that the voltage doubled supply from the 5V winding is generating a negative power supply to the bottom of that CCS circuit. Can't recall the exact voltage (and anyway its not that critical) but its about -12Volts. Cheers, Ian