Integrated Amp: Mark III & PAS

for Dynaco Mark II/III/IV and DIY PP monoblocks

Postby ChrisK » Fri May 15, 2009 10:05 am

Blair wrote:I have heard both, and I prefer SS rectification personally. It is all subjective. Here is the amps I keep referring to even though it is before the 70v bias tap was available. It is the AN-4T400. I used a separate tranny to get my bias, but your tranny will be fine.
Blair


Blair, do you recall the source for the extruded toroid cover that you used? I've been googling all over the place but find only manufactures, not retailers...

Thanks, Chris
User avatar
ChrisK
 
Posts: 115
Joined: Sun Mar 23, 2008 5:57 pm
Location: Laguna Beach, CA

Postby Blair » Fri May 15, 2009 10:09 am

LOL....You will laugh at this, but it is a utensil can from Target! They do not sell one that fits over the toroid you are buying though. Your best bet is speedymetals.com select 6" OD aluminum tube with 1/8" walls. Then top it with a piece of plate.

I cut the can down and then trimmed the bottom edge with automotive door trim for a nice finished look.

I may have missed this, but are you building a pair of amps or a stereo amp?

Blair
Blair
KT88
 
Posts: 1272
Joined: Thu Nov 09, 2006 10:22 am

Postby ChrisK » Fri May 15, 2009 11:11 am

Blair wrote:LOL....You will laugh at this, but it is a utensil can from Target! They do not sell one that fits over the toroid you are buying though. Your best bet is speedymetals.com select 6" OD aluminum tube with 1/8" walls. Then top it with a piece of plate.

I cut the can down and then trimmed the bottom edge with automotive door trim for a nice finished look.

I may have missed this, but are you building a pair of amps or a stereo amp?

Blair


Yessiree, that's the kind of stuff I like to hear! Innovation is the heart of invention or something like that...

I have no problem with metal work. In fact I have aluminum plate in my workshop that I use to build home-made solar panels.

I'm building a stereo amp on a single chassis to save space: 2 Poseidon driver PCB's, new caps and resistors, Mark III A-431's, the toroid that will reduce chassis width by some 3" and weight by 9 lbs. Also, I've researched Mark III's on the net for two years and have a great reference collection from all the folks out there sharing their (credible) improvements.

Generally, my goal is to build decent amps using good, readily available components. I'm not the kind of guy that pays $50 for "the World's Best" paper in oil signal capacitor 8-)

Does your "targay can" run cool or does it need vent holes?

Chris
User avatar
ChrisK
 
Posts: 115
Joined: Sun Mar 23, 2008 5:57 pm
Location: Laguna Beach, CA

Postby Blair » Fri May 15, 2009 11:22 am

Hey Chris,

It runs pretty cool. The heat is from the heater windings. I have another pair of BIG!!! amps using a separate tranny for the filaments and the toroids run very cool.

If I remember correctly, your toroid will have 9A of heater winding available That should be sufficient for 4 KT88s and your three driver tubes assuming you use the ST-70 board instead of the Poseidon boards to save space.

Blair
Blair
KT88
 
Posts: 1272
Joined: Thu Nov 09, 2006 10:22 am

Postby ChrisK » Tue May 19, 2009 7:55 pm

TerrySmith wrote:Instead of two seperate P-782 trannies, I would use one toroid such as the Antek 4TK-400, it'll power all of it with no problem. Plus you won't need a forklift to move it around.


Well, the Antek 4TK-400 arrived and there's good news that might be useful to others.

The model I received is wound with a 5v tap for the 5AR4/GZ34, two 6.3v taps and two 400v secondaries with a 70v bias tap each. The builder now has a choice between tube or SS rectification all with one toroid.

Time to draw up the schematic.

Chris
User avatar
ChrisK
 
Posts: 115
Joined: Sun Mar 23, 2008 5:57 pm
Location: Laguna Beach, CA

Postby Blair » Tue May 19, 2009 8:29 pm

Chris,

Be careful! I think that 6.3v tap you refer to is center tapped. I have that tranny on these:

Image

See familiar colors ;)

You should have 800vct, 6.3vct@5-6A, 70v, and 5v

Just be sure you are not trying to power your filaments with 3.15v and expect to get enough heat.

Blair
Blair
KT88
 
Posts: 1272
Joined: Thu Nov 09, 2006 10:22 am

Postby ChrisK » Tue May 19, 2009 10:13 pm

Blair wrote:Chris,

Be careful! I think that 6.3v tap you refer to is center tapped.

You should have 800vct, 6.3vct@5-6A, 70v, and 5v

Just be sure you are not trying to power your filaments with 3.15v and expect to get enough heat.


Thanks for the heads up! Yes, it's a center tap for 3.15v, I see that now and I'll have to correct my drawing.

At the risk of showing my ignorance, what kind of a circuit/device would use the 3 volts? Must be a reason for it.

I like the pic: Looks like your wire discipline keeps getting better 8-)

Chris
User avatar
ChrisK
 
Posts: 115
Joined: Sun Mar 23, 2008 5:57 pm
Location: Laguna Beach, CA

Dual 5AR4's

Postby ChrisK » Wed May 20, 2009 10:41 am

Sure could use a quick sanity check here.

I'm building two Mark III amps on one chassis using a single toroid power transformer that has two discreet 400v outputs (no CT) and one 5v tap for the 5AR4 heaters. Everything went well until I realized that the heater interconnects ended up also tying the two B+ outputs together.

Would this work or am I setting up a meltdown here? If not, I could always get a second 5v heater source.

Image

Thanks in advance,

Chris
User avatar
ChrisK
 
Posts: 115
Joined: Sun Mar 23, 2008 5:57 pm
Location: Laguna Beach, CA

Postby EWBrown » Wed May 20, 2009 11:14 am

The main requirement is tha the5 VAC winding has enough current capability to light up two 5AR4/GZ34s (4 amps minimum).

That circuit, as it is shown, would not work, as there is no "return" for the negative portion of the AC voltage. Since the two secondaries do not have center taps, there are two approaches to make this thing work properly:

First, is to connect the secondaries in series, so it looks like one 400-0-400 secondary, ground the center tap, and you can parallel the two rectifier tubes in order to handle the higher current.

The advantage is simplicity, the disadvantage is that there will be only one B+ supply.

Second, which is IMO the preferred method, is to start out with your circuit, and then add two SS rectifier diodes to each secondary, anodes connected together and to ground (zero volts) bus, and connect one SS diode's cathode to each of the rectifier tube plates (pins 4 and 6).

This method really should have two separate 5VAC filament windings, read on...


This will give you a "hybrid" full wave bridge, with the advantage of the tube rectifier's slow warmup and soft atart, and the two SS diodes will handle the negative half of the secondaries' AC voltage.

Now, in order to take full advantage of this, you will need another 5VAC transformer, so that each rectifier's cathode / filament is separate, in order to get two separate B+ suplies.

It is conceivably possible to simply add a second 5VAC winding, with properly insulated wire, wound around the outside of the toroid's existing windings.

You would need to know the number of turns per volt, which is a fairly simple procedure to calculate - just wind ten (or any convenient number) of turns, and measure the voltage (under about 2 amp load). The actual number of turns is generally between 5 and 10 turns per volt.

Since the transformer will have lorts of "VA" reserve, this won't over-drive the primary windings' current capabilities.

This will add 10VA to the existing power consumption, which is about 2.5% of the 400VA trannie's overall capabilities. Pretty trivial....

This also allows you to have two separate B+ supplies, one for each channel, so you will have a "dual mono" DC power distribution.

BAck to your other question:

The 6.3VAC filament winding's center tap is just a handy grounding reference - or connect it to a DC "bias" in order to raise the filament's potential above the tubes' cathodes - IIRC, the Mark III uses a capacitor from the filament CT to ground, this allows the filaments to "float" a at a level above ground potential, whhich is a cheap and simple way to reduce any 60Hz AC hum problems.

HTH

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby ChrisK » Wed May 20, 2009 12:21 pm

Thanks very much for your quick and thorough response. You've helped me immeasurably, especially by introducing the concept of the "hybrid tube/SS" rectification scheme.

EWBrown wrote:That circuit, as it is shown, would not work, as there is no "return" for the negative portion of the AC voltage.


Yup, I realized that right after I sent my post. No return, no current flow.

EWBrown wrote:IMO the preferred method, is to start out with your circuit, and then add two SS rectifier diodes to each secondary, anodes connected together and to ground (zero volts) bus, and connect one SS diode's cathode to each of the rectifier tube plates (pins 4 and 6).



IMHO, your suggestion is spot-on because it resolves so many real and perceived issues all in one swoop: inrush limitation, cathode stripping, no need for a complicated "soft-start" circuit, and maintains an elegant simplicity.

I've spent dozens of hours researching the SS rectification B+ issue in tube amps but have never come across this solution before. Just yesterday, before you suggested the hybrid bridge, I concluded that the guitar amp builders might have the right idea: a simple standby switch that cuts B+, although I had some reservations about the safety of running a metal toggle switch 500% above it's voltage rating.

Alternatively, I was just starting to look into the Amperite delay tubes.

I'm at work now (kind of), but I'll revise the circuit later today to illustrate your concept and post it in a follow-up message. I'd appreciate your quick "once-over" to make sure it's what you intended.

EWBrown wrote:It is conceivably possible to simply add a second 5VAC winding, with properly insulated wire, wound around the outside of the toroid's existing windings.


Another good idea. Sounds like I get to experiment this weekend and see what the VTVM tells us. If it looks promising, I'll buy some enameled wire and give it a go.

Thanks very much for sharing your expertise with us.

Regards,

Chris
User avatar
ChrisK
 
Posts: 115
Joined: Sun Mar 23, 2008 5:57 pm
Location: Laguna Beach, CA

Postby ChrisK » Wed May 20, 2009 8:08 pm

EWBrown wrote:IMO the preferred method, is to start out with your circuit, and then add two SS rectifier diodes to each secondary, anodes connected together and to ground (zero volts) bus, and connect one SS diode's cathode to each of the rectifier tube plates (pins 4 and 6).

This method really should have two separate 5VAC filament windings


If I read correctly, I think this is what you described:

Image

I have some 1N5008's lying around that are rated to 1000v. Would there be any point in tying a resistor in parallel with each diode? Forgot why, but I've seen it done on some of the circuits I've studied.

Thanks again, this is great!

Chris
User avatar
ChrisK
 
Posts: 115
Joined: Sun Mar 23, 2008 5:57 pm
Location: Laguna Beach, CA

Postby EWBrown » Thu May 21, 2009 7:18 am

For the added 5VAC winding, the insulation should be able to withstand at least 600V, and I don't know what is the insulation voltage rating of enameled "magnet" wire.

If you were to add an extra insulating wrap over the added windings, then perhaps that would be sufficient.

Otherwise just an additional 5VAC filament trannie under the chassis would do the trick.

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby EWBrown » Thu May 21, 2009 7:22 am

You new schematic is right-on!

Rather than having a resistor in parallel with the SS rectifiers, perhaps a 10 ohm, 1W resistor in series with each diode would help absorb some of the transients, or just skip the added resistors altogether. The tube half of the hybrid bridge pretty much eliminates the transient spike situation, anyway.

A resistor in parallel would only serve to inject unwanted AC onto the DC, and also would carry the full reverse voltage on the opposite half-cycle, so they would eventually burn out. .

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby EWBrown » Thu May 21, 2009 7:24 am

Are the 70V bias taps a separate winding, or are they taken off the 400V windings?

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby ChrisK » Thu May 21, 2009 7:39 am

Good morning,

EWBrown wrote:Otherwise just an additional 5VAC filament trannie under the chassis would do the trick.


Adding windings to the toroid properly is iffy so I'll just buy a little 5v trannie. Hammond has a little 2.5v CT 3a for about $18. Sounds like a good value considering the time spent on the alternative even though it sounds fun.

Chris
User avatar
ChrisK
 
Posts: 115
Joined: Sun Mar 23, 2008 5:57 pm
Location: Laguna Beach, CA

PreviousNext

Return to poseidon

Who is online

Users browsing this forum: No registered users and 16 guests