'AK Mods' SC Class A 15W Amplifier

[jeffc]

I could be wrong, but it took a great deal of experimentation, involving similar phasing of both primary and secondary windings of both transformers to achieve the present results, which also resulted in improved channel separation as well.

Bloody hell Alex,

You like to rub in the fact that my 2 of the 4 'on sale' transformers have almost random primary and secondary wire colour sequences, so phases are a best guess

Replies from TorTech on my Qs about the transformers
1.
"Hi Jeff,
Sorry about that.
As the unit is centre potted, mounting kits are not usually used on these units, but we will send you the normal mounting kits if you wish.
In regards to the leads, while the lead out sequences may differ, the colour codes are the same.
Primary is 2*115V Blue-Yellow, Orange-Brown.
Secondary is 2*20V Red - Green, White - Black."

2.
"Jeff,
Ok, let me know if you want any mounting kits. The potted unit does have a pad underneath, and only a centre bolt is required to panel mount it, although a top plate and pad can still be used also.

While it is unusual for the leads to be in different sequences, as is said, the colours are still correct, no matter what the position.
Having said that, of course, I cannot rule out that there is a fault here, unless you try it out, or return them for checking, however I have checked two batches here, that are correct, so I am confident.

The colours and Phase are in the order written below, and I have checked a stock unit here.
So 230V is Blue - Brown, with Orange to Yellow joined, Output 40V is Red - Black, with Green to White joined."

Based on the pic of wiring with these transformers, would you guys be contented that primary/secondary phases will be correct based on these responses?

Tempted to hook up the winding pairs in parallel based on phases suggsted using a 2-3A fuse on input active, as if windings are connected in anti-parallel, according to Rod Elliott, the fuse will certainly let me know about this.



cheers.. jeffc

[Deleted]

You like to rub in the fact that my 2 of the 4 'on sale' transformers have almost random primary and secondary wire colour sequences, so phases are a best guess

Hi Jeff
Feeling picked on ?
The Primary phasing was based on my own experiences with the SC DAC, where one of the original 30VA transformers had 2 orange wires.
The secondary wire phasing was based on my own experiences with the Class A HA !
The wire colour sequence that were quoted are usually printed on the side of the transformer as in the attached diagram
The 2 primary centre wires are joined for 230V AC .
The 2 centre secondary windings are joined for centre tap operation.

Blue and orange, and yellow and brown would have been connected together for 115V AC.
Red and white, and green and black would have been connected together for 20V AC at double the current.
So yes, what he has said does appear to be correct.

Tempted to hook up the winding pairs in parallel based on phases suggsted using a 2-3A fuse on input active, as if windings are connected in anti-parallel, according to Rod Elliott, the fuse will certainly let me know about this.

Why on earth would you want to do that ?
Can't you get into enough trouble with SWMBO without doing that ?
Regards
Alex
P.S.
It is far more satisfying to put an electrolytic in the wrong way around, than to blow a fuse! ;D

Attachments:

[Will]

From a quick read, I think that this link may have some useful info for you, Jeff, and perhaps anyone that needs to establish true phase for their traffos.

nptel.iitm.ac.in/courses/IIT-MADRAS/Electrical_Machines_I/pdfs/1_7.pdf

In a bit of a rush, so I may have missed something.

[Deleted]

It is far more satisfying to put an electrolytic in the wrong way around, than to blow a fuse!

I've managed to get an Oscon to go off like a friggin' bullet, very impressive!

Jeff,

I followed Alexs recommendation on the transformers in the SCDAC, I too originally had one with two orange primary leads.
Being a cynical sod, I had to try adding only one new TX with coded primaries first. I obviously, by fluke, had it right first time as on swapping the orange primaries around the SQ deteriorated slightly. That allowed me to mark them before then changing for two, same model, TXs with coded leads.

It's a bit poo that your TXs are not marked for connectivity.

[jeffc]

Thanks guys,

Alex, indeed I sent this same schematic to TorTech to try and get one sent back specifically for the Nuvotem transformers I bought. They mustn't have anything.

nptel.iitm.ac.in/courses/IIT-MADRAS/Electrical_Machines_I/pdfs/1_7.pdf

Will, exactly what I need , a test method to ensure I can wire these wacko transformers identically for each amp board, excellent find, you've saved me some sleepless nights worrying

cheers.. jeffc

[Deleted]

Dec 6, 2011 23:37:42 GMT jeffc said:

Thanks guys,

Alex, indeed I sent this same schematic to TorTech to try and get one sent back specifically for the Nuvotem transformers I bought. They mustn't have anything.

nptel.iitm.ac.in/courses/IIT-MADRAS/Electrical_Machines_I/pdfs/1_7.pdf

Will, exactly what I need , a test method to ensure I can wire these wacko transformers identically for each amp board, excellent find, you've saved me some sleepless nights worrying

cheers.. jeffc

Hi Jeff
Bloody hell, you will blind yourself with science !
Just keep the top 2 in the photo for yourself, use the methods of connection that I have already posted, that confirm what the maker said, AND GIVE THE OTHER 2 TRANSFORMERS TO PHIL ! ;D
Alex
P.S.
The bottom 2 seem to show a manufacturing inconsistency, or perhaps put together by a different employee.

[jeffc]

Just keep the top 2 in the photo for yourself, use the methods of connection that I have already posted, that confirm what the maker said, AND GIVE THE OTHER 2 TRANSFORMERS TO PHIL !

You've obviously attended the 'How To Win Friends And Influence People' course, NOT

P.S.
The bottom 2 seem to show a manufacturing inconsistency, or perhaps put together by a different employee.

Albert Einstein moment , yer one from outer space.

cheers.. jeffc

[jeffc]

Using a old 10V AC wall wart

Primary is 2*115V Blue-Yellow, Orange-Brown.
Secondary is 2*20V Red - Green, White - Black."

primaries Blue - Yellow, Orange - Brown (~11.5V ea = V1)
secondaries Red - Green, White - Black (~2.5V ea = V2)

Primary to secondary voltages
Blue - Black or Blue - Green (~9.5V = V3)
Blue - Red or Blue - White (~7.2V = ?)

Same voltages obtained with the Orange primary

Voltages with the other colour wire of each primary connected to secondaries all low (~2V)

So for... V1 - V2 = V3 is 11.5 - 9.5 = 2.5

Thus
2 x Primary windings start (dot): Blue and Orange (as specified)
2 x Secondary windings start (dot) Green and Black (arse about than was specified but will still have White-Green tied together for 0V as specified, so I could have just followed instructions ).

Easy method Will, top stuff , and as its nice and cool here tonight (indeed a record low for a Dec max temp of 19C), and with these pesky transformers now sorted, I can sleep easy.

cheers.. jeffc

[Deleted]

Hi Alex,
I'm scrubbing that idea of using the two positive regs for the PSU's +/- rails having taken on board your comments.
Also Frans put me right on a few other points for which I'm very great-full.
However the whole exercise was very worthwhile and jangled my old grey matter around a bit which is all to the good. ;D

So what next.......................
well I've added the CL section to my combo PSU/JLH and will use the LT1085/LT1033 to get the +/- 20volts just need to get some LT1033s ordered and a few other bits.
Not made a decision on the rectifier section yet might just go with one of those potted versions used in the original SC PSU
Santa arrived to day with my Airlink traffos courtesy of my doting boys............. that'll get me started on a test bed to set them up in which I hope will end up as either the traffo box or a section of the whole thing.
Regards,
Alan

[Deleted]

Hi Alan
That will make it easier for long distance diagnosis by Will and myself, in the event that you have minor niggly problems, or something sounding below expectations.
I am likely to try the new PSUs in mine, but using a single Panasonic 10,000uF 35V in each rail, with the 3.5A 150V fast, soft recovery diodes.
Regards
Alex

au.element14.com/panasonic/ecos1va103ca/capacitor-10000uf-35v/dp/1198694
australia.rs-online.com/web/p/products/6393308/?searchTerm=639-3308&relevancy-data=636F3D3126696E3D4931384E525353746F636B4E756D6265724D504E266C753D656E266D6D3D6D61746368616C6C26706D3D5E5C647B337D5B5C732D2F255C2E5D5C647B332C347D2426706F3D313426736E3D592673743D52535F53544F434B5F4E554D424552267573743D3633392D333330382677633D4E4F4E4526

[jeffc]

After all my transformer hassles, I'm thinking of trying these custom set for +/-20V
www.connexelectronic.com/documents/SMPS300R.pdf

[Deleted]

Hi Jeff
You wouldn't do that! ;D
Kind Regards
Alex

Attachments:

[Deleted]

Dec 8, 2011 1:40:36 GMT @sandyk said:

Hi Alan
That will make it easier for long distance diagnosis by Will and myself, in the event that you have minor niggly problems, or something sounding below expectations.
I am likely to try the new PSUs in mine, but using a single Panasonic 10,000uF 35V in each rail, with the 3.5A 150V fast, soft recovery diodes.
Regards
Alex

au.element14.com/panasonic/ecos1va103ca/capacitor-10000uf-35v/dp/1198694
australia.rs-online.com/web/p/products/6393308/?searchTerm=639-3308&relevancy-data=636F3D3126696E3D4931384E525353746F636B4E756D6265724D504E266C753D656E266D6D3D6D61746368616C6C26706D3D5E5C647B337D5B5C732D2F255C2E5D5C647B332C347D2426706F3D313426736E3D592673743D52535F53544F434B5F4E554D424552267573743D3633392D333330382677633D4E4F4E4526

Hi Alex

just a quick question

i noticed that you intend to bump the input cap up to 10.000uf per rail which is more than twice the size of the original design (4700uf).

I'm pretty sure that's not a light minded decision so I'm wondering what you reasoning is on that?

take care

[Deleted]

By, or not by, coinicidence, the designer of my old integrated also recommends an increase from 4 x 4700uf to 4 x 10000uf. Stating better bass as the main gain.

I only went up a couple of steps but liked the result. A certain ease of flow becomes apparent too.

(Physical size of the caps at the time was the only restricting factor in the original design.)

[Deleted]

i noticed that you intend to bump the input cap up to 10.000uf per rail which is more than twice the size of the original design (4700uf).I'm pretty sure that's not a light minded decision so I'm wondering what you reasoning is on that?

Hi Shaun

The original 5,600uF are more bulky, and more expensive.
I can get Panasonic 10,000uF 35V from Element 14 for just over AU$6 each.It certainly isn't normally needed, but seeing we will also have greater current capacity due to 2 toroidals, and twin PSUs, it may help those who use 4 ohm speakers, where if pushed harder it will go into Class AB for up to 30W on peaks.
Regards
Alex

[Deleted]

Dec 10, 2011 0:48:21 GMT @sandyk said:

i noticed that you intend to bump the input cap up to 10.000uf per rail which is more than twice the size of the original design (4700uf).I'm pretty sure that's not a light minded decision so I'm wondering what you reasoning is on that?

Hi Shaun

The original 5,600uF are more bulky, and more expensive.
I can get Panasonic 10,000uF 35V from Element 14 for just over AU$6 each.It certainly isn't normally needed, but seeing we will also have greater current capacity due to 2 toroidals, and twin PSUs, it may help those who use 4 ohm speakers, where if pushed harder it will go into Class AB for up to 30W on peaks.
Regards
Alex

Hi Alex

thanks for the reply

as luck would have it i have a stash of 10.000uf caps in my spares box so I'll keep some handy.

my speakers are nominally 6ohm so a little more in the tank may help those along.

take care

[Deleted]

Hi Alex,
Farnell UK have these caps @ £3.45 tax incl. that's £1 sterling cheaper than the Au price.

RS UK are £6.23 for the same cap..............whose got it wrong here?
I'll set my board spacing for 25mm dia' caps and hope the dust settles before the price goes up.

With regard to total capacitance, Frans has this to say

When powering up the circuit 20.000uF is a complete short momentarily and since a 160VA trafo is used HUGE currents will flow.
The suggested diodes are not designed for this.
The bridge rectifier is.

Alex and other subjectivists will certainly disagree

Regards,
Alan

[Deleted]

Dec 10, 2011 14:14:43 GMT drymdrum said:

Hi Alex,
Farnell UK have these caps @ £3.45 tax incl. that's £1 sterling cheaper than the Au price.

RS UK are £6.23 for the same cap..............whose got it wrong here?
I'll set my board spacing for 25mm dia' caps and hope the dust settles before the price goes up.

With regard to total capacitance, Frans has this to say

When powering up the circuit 20.000uF is a complete short momentarily and since a 160VA trafo is used HUGE currents will flow.
The suggested diodes are not designed for this.
The bridge rectifier is.

Alex and other subjectivists will certainly disagree

Regards,
Alan

Hi Alan

i was wondering about the devils own soft start circuit if i end up using the bigger caps but I'm not sure it's needed.

i have one to hand just in case but i thought that was the way round Frans's point about inrush current.

worth talking about? maybe

I've looked back through the posts on this thread and can't find the post that the above two quotes come from ?

subjectivist,objectivist can't we just be music lovers?

it's that sort of name calling that stifles discussions and is just not helpful.

take care

[Deleted]

Alan and Shaun
I would not recommend any more than 10,000uF per voltage rail.
The 3.5A diodes that I suggested, and the larger TO220 diodes that Will mentioned, are well up to these requirements without any kind of soft start circuit. For those who do not have 4 ohm nominal loudspeakers, 4,700uF or 5,600uF is fine. It is already double that of the original design which used 5,600uF per rail in a supply for BOTH PCBs. I will be using 10,000uF this time so that I can perhaps do some listening tests into 4 ohm speakers in order to be across any possible problems those of you with 4 ohm speakers may encounter.
Don't forget to use individual 3A fuses for each transformer's primary windings.
Regards
Alex

P.S.

Quote: When powering up the circuit 20.000uF is a complete short momentarily and since a 160VA trafo is used HUGE currents will flow.
The suggested diodes are not designed for this.
The bridge rectifier is.



Quote: Alex and other subjectivists will certainly disagree

I am at a loss to understand this one, as each PAIR of the 3.5A diodes ( or higher rated TO220 types) will charge only the 10,000uF capacitor associated with it, where a bridge rectifier configuration is used. Perhaps a different circuit was being discussed here ?

[Deleted]

Hi Alex ,
Frans was commenting on the circuit which I was proposing to use as PSU for the 15 Watt......Previously posted in this thread.......
This incorporated the use of two POSITIVE REGS. LT1085 from which the Pos/Neg supply would be derived. Having taken on board your doubts as to the suitability of this circuit I ran it past Frans and his observations on top of your own were enough to show me the error of my ways.
As you know I have reverted to the use of LT1085/33 for Pos/Neg supply in the configuration used in the original design.
Many thanks to both Frans and yourself,
Regards,
Alan

[Deleted]

Hi Shaun,
quotes are from a PM from Frans to me you wont find them posted.
Perhaps I was wrong to post from a PM.......not sure on the etiquette of that..................so my apologies to you Frans .
However we, ......that's you and I Shaun......have our reply from Alex which helps us make a more informed decision on how to populate our PSU.............that has to be good.
As to the.....SUBS and OBS thing'y I doubt if any RG member is wholeheartedly one or the other .................in Frans case I have always viewed his use of the word......... "Subjectivist" with regard to Alex............. more a term of endearment than anything else .
Take care,
Alan

[Deleted]

HI All,
got my final layout for PSU done.....I hope.
It stuffs up fine on the foamboard measures 150mm X 50mm
and combines the LT1085/33 Pos/Neg PSU with Pos/Neg JLH.
I'll give it a full eyeball over the next few days then carve....mill...rout....cut a test board.
So more to come on that score.
Meanwhile here's a couple of pics to be going on with
Alan

[Deleted]

This image has been modified by drymdrum

[Deleted]

This Image has been updated by DrymDrum and is now at Version 4

[Deleted]

Alan
Where does a fellow 70+ year old member get so much energy and drive from ?
Kind Regards
Alex