DIY Audio Projects Forum
http://diyaudioprojects.com/Forum/

300B SET Design Project
http://diyaudioprojects.com/Forum/viewtopic.php?f=9&t=2699
Page 9 of 53

Author:  mwhouston [ 23 Oct 2010, 16:49 ]
Post subject:  Re: 300B SET Design Project

Suncalc wrote:
Hmmmm....

This is a little fly in the ointment. I was just getting ready to post the final PS design. It is based on an Edcor XPWR077. But that's not what the customer wants. :bawling:

So... For the output transformers, thats easy. Go with the Hammond 1630SEA. It meets all requirements and we already considered it in our design. For the primary PS choke, I've already settled on the Hammond 193J 10H choke (or the 193M or 193Q if you can't get the 'J').

The power supply transformer is another matter. Hammond only makes two transformers that meet the PS requirements for this amp. These are the 282X or 382X ( 1000v(500-0-500)@200mA, 5v@3A C.T., 6.3v@6A C.T.). And I couldn't find either of these on the web page you linked. If I need to go to the lower voltage (400-0-400) transformer it will mean an entirely new design from the ground up.

Please let me know what you want to do.

Sorry to stuff you around but after more talks with the client we can use Edcor power tranni just Hammond OPTs. Hope this makes it easier.

I have just e-mailed my local Hammond distributor to see if they stock the power trannies you mention.

Author:  Suncalc [ 23 Oct 2010, 17:54 ]
Post subject:  Re: 300B SET Design Project

This makes things a lot easier. :D

I'll go with the Edcor XPWR077 for the power transformer and use the Hammond 193J 10H filter choke. Go ahead and order the 1603SEA output transformers and we should be set. I am also planning on using the Hammond 266K12 dual-primary/dual-secondary transformer for the 300B DC filament supplies. So just to be clear on the iron, here's the list:

Power Transformer: Edcor XPWR077 900V(450-0-450)@200mA, 6.3V@5A & 5V(2.5-0-2.5)@5A
Filter Choke: Hammond 193J 10H (200mA, 82Ω)
Filament transformer: Hammond 266K12 "Dual Primary/Dual Secondary" 18.9VA 2x6.3v@1.5A
Output Transformers: Hammond 1630SEA 3500Ω 30W 135mA Single Ended (x2)

This should cover the whole amp for iron.

A comment about the filter caps. I understand the desire to use motor run caps for filtering. The low cost and ready availability make them seem an attractive option. However in my experience these motor caps tend to be noisy in audio circuits. This is definitely a component construction issue with the capacitor design being optimized for other parameters. I'll post my power supply design in a little while. But I'll tell you up front the capacitors called for are 40µf, 100µf, and 47µf all at 500v minimum.

Hope this helps.

Author:  Suncalc [ 23 Oct 2010, 18:44 ]
Post subject:  Re: 300B SET Design Project

Sampleaccurate;

Understand about the 1627 vs. the 1630. Here is the load line for the 2.5K 1627SEA using the same bias point.

Attachment:
2p5k load line.jpg


Outside of the slightly higher power and slightly higher distortion, this looks all right. However, if you look at the Hammond specifications you'll notice that the effective inductance of the 1630 is more then twice the inductance of the 1627 (42H vs. 20H). This higher inductance equates to much higher stored energy in the core from the bias current. This higher stored energy will definitely have an effect on the responsiveness of the amplifier.

It would be very interesting to to compare the 1627SEA with the secondary double loaded as you suggest (8Ω load on the 4Ω tap) with the 1628SEA single loaded in the same amplifier. I am willing to bet the increased inductance of the 1628 would make a significant difference. Unfortunately it's kind of an expensive experiment. :$: :$:

Author:  sampleaccurate [ 23 Oct 2010, 19:38 ]
Post subject:  Re: 300B SET Design Project

Suncalc wrote:
Sampleaccurate;

However, if you look at the Hammond specifications you'll notice that the effective inductance of the 1630 is more then twice the inductance of the 1627 (42H vs. 20H). This higher inductance equates to much higher stored energy in the core from the bias current. This higher stored energy will definitely have an effect on the responsiveness of the amplifier.

It would be very interesting to to compare the 1627SEA with the secondary double loaded as you suggest (8Ω load on the 4Ω tap) with the 1628SEA single loaded in the same amplifier. I am willing to bet the increased inductance of the 1628 would make a significant difference. Unfortunately it's kind of an expensive experiment. :$: :$:


I noticed the inductances (42 vs. 20 Henrys for the 1630 and 1627 repectively) and also the 1628SEA (a 5K, 48 Henry primary) and wondered exactly the same thing - what would be the sonic difference between a 1628SEA at 5K using the correct tap vs. the 1627SEA using the 4 ohm tap on an 8 ohm speaker for a reflected impedance of 5K? Same reflected impedance but 48 Henrys vs 20 Henrys. That's one I'd have to think about.

My initial reaction would be that the 1628SEA would perform better having a higher inductance and consequently lower current required to magnetize the core. I believe it's a trade off of higher flux leakage but lower magnetizing current for a high inductance tranny vs. a lower magnetic leakage but also a higher magnetizing current and less efficient energy transfer for low inductance trannys. I'm not sure what the sonic difference would be. I'm not an expert on OTs. Maybe one day.

Author:  mwhouston [ 23 Oct 2010, 20:12 ]
Post subject:  Re: 300B SET Design Project

Matt: I don't have to use motor starts. Let's go Mundorf all the way. Thanks for the iron specs. I can now order.

Author:  Suncalc [ 23 Oct 2010, 22:21 ]
Post subject:  Re: 300B SET Design Project

Mark!

I hope you haven't ordered yet. I have been finalizing the power supply design and I think I need to call out the Hammond 193M instead of the 193J filter choke. The design really needs the lower 63Ω resistance. So the choke you need is...

Filter Choke: Hammond 193M 10H (300mA, 63Ω)

I hope this doesn't cause any problems.

Author:  mwhouston [ 23 Oct 2010, 22:39 ]
Post subject:  Re: 300B SET Design Project

No order yet.

Author:  mwhouston [ 24 Oct 2010, 07:05 ]
Post subject:  Re: 300B SET Design Project

Matt: I talked of going Mundorf all the way with filter caps. These are the MTube caps and appear to be the best they have.

They don't "do" a 40uf. So what would you prefer: 30uf or 47uf. I'm about to order.

http://www.partsconnexion.com/capacitor ... mtube.html

Author:  Suncalc [ 24 Oct 2010, 10:35 ]
Post subject:  Re: 300B SET Design Project

Mark;

Since there won't be any hot switching on the 5U4 lets go with the 47µf and I'll work the current in rush issue separately.

With the 40µf input cap the wRC is already only at 35 and the output voltage is just at the level required. Going to a 30µf will drive wRC down to 26 which is too low to get the required output voltage. I'm not that worried about overall regulation because the PS load for a SET is fairly constant but I am concerned about transformer efficiency (to avoid too much transformer heating) and overall output voltage. Attached below is my PS design sheet for the 40µf input cap. It meets the target ripple spec and voltage, the higher input will give a better ripple number :up: but any increase in voltage will probably be eaten up in transformer series resistance to keep the rectifier inrush current at a reasonable level.

Attachment:
PS Load Sheet 40.jpg


I'll rerun the numbers with the 47µf and get the design posted. I am fairly confident the the 47µf will be ok.

Author:  mwhouston [ 24 Oct 2010, 15:51 ]
Post subject:  Re: 300B SET Design Project

That would be great I can order all the filter caps then.

I'm waiting on Evatco, my local Hammond dealer, to be back to me about the OPTs. If they can't supply I'll have to convince the customer that the Edcors are OK. Which I have already done once.

Does Edcor have the a suitable filement tranni? I'm not sure my local Hammond store stocks them?

Look forward to your final calcs with 47uf first cap. Also I am thinking about a remote PS for the amp. Power tranni, choke etc. on a seperate base to 300Bs, 6SN7 and OPTS. I understand 300Bs can easily pick-up AC.

Page 9 of 53 All times are UTC - 6 hours [ DST ]
Powered by phpBB® Forum Software © phpBB Group
http://www.phpbb.com/