Hi John, That's magnificent you've set the bar pretty high there for us mortals.

I'm still lurking, listening and learning thanks for the inspiration

Quick questions does the Golden Reference supply make an audible difference? I know that's subjective but I haven't ordered them and maybe too late.

What is your Transformer? It looks really tidy.

Cheers

Neil

Hi Neil,

I think the difference is subtle but worthwhile, the GRLV offers a slightly more stable presentation. The transformer is from Sumr, fully encapsulated as well as core band shielding.

John

Edited September 6, 20169 yr by johnwmclean

Wow John!  Your carbon looks amazing!  Thanks for posting those pictures.  Gotta love the RK50 in there.  Gotta be the best build around!

Beautiful build John! 

About GRHV. BC557 has -45V voltage limitation. If you use a way over spec trafo and unluckily enough to turn on the amp on the peak of AC, before the regulation starts working, the voltage crossing BC557 will be over 45V. This might be a long term problem. Or, when sth is wrong when assembling HVs, BC557 will be easily fried.

2 hours ago, joehpj said:

About GRHV. BC557 has -45V voltage limitation. If you use a way over spec trafo and unluckily enough to turn on the amp on the peak of AC, before the regulation starts working, the voltage crossing BC557 will be over 45V. This might be a long term problem. Or, when sth is wrong when assembling HVs, BC557 will be easily fried.

I'm not sure I see this.  The BC557 are part of a current mirror for the error amp.  One of them is bound to about .6V and the other I see peeking at around 7V in the simulation.

I'm not sure what the AC or an overly spec'd transformer has to do with this.  Also there is a natural slow start in the regulator so it comes up under 1V until it starts regulating.

Are you actually measuring these values?  I'd be curious if that's true.  Reality and simulations don't always align.

There are some conditions if the circuit is not put together correctly that they could fry but that's true of a number of parts.

A friend was building this amp and said something was wrong in the bias of 1000170D he's building, causing the regulation not working. Thus, the output voltage was around 457V. Few seconds later, the BC557 fried. If BC557 was substituted with some higher voltage sands (~100V), the circuit will be more fail-proof.

That's possible, but they should find the underlying issues as well.  I use 60V parts in my builds, though my prototypes used BC557.

deleted
/all-ready answered

Edited September 6, 20169 yr by sorenb

John, that is a truly enviable piece of gear.  One of few amps that make an RK50 look like a reasonable volume control.

Wow wished i had the kind of dedication to build a carbon, and to build one like johnwmclean . Totally epic build quality! If that ever got posted for sale.. would sell.. in under 2 seconds.

Is there any TOSHIBA substitute for 2SK170? Because this hard to buy LSK389/170 here.

Some say 2SK366 is ok. How do you guys think?

On 9/4/2016 at 4:03 PM, spritzer said:

My voltage selector board. 

Where can I get more info on this board?

For dual primary transformers it uses this simple selector switch;

http://au.mouser.com/Search/m_ProductDetail.aspx?R=V80212MS02QEvirtualkey61170000virtualkey611-V80212MS02QE

Has extra terminals for grounding and AC switch capability.

It's very simple.

41 minutes ago, johnwmclean said:

For dual primary transformers it uses this simple selector switch;

http://au.mouser.com/Search/m_ProductDetail.aspx?R=V80212MS02QEvirtualkey61170000virtualkey611-V80212MS02QE

Has extra terminals for grounding and AC switch capability.

It's very simple.

Thanks. But I meant where could I get the PCB?

I used to sell them but I don't do so any more. 

Are Gerbers available? I don't mind making a bunch because almost all my projects need international compatibility.

No gerbers will be released for this one. 

I need some advice with the construction of a power cord.

I am ready to case my Carbon finally and it will be a two-chassis layout; one chassis for the power supply (GR HV, LV and transformers) and one chassis for the amp. The power supply connection will be through a 7-wire cable (ground, +580 VDC, +400 VDC, -400 VDC, +15VDC, -15VDC and one wire will be unused for now). I plan for the cable to be about 3 feet long to provide flexibility in placement.

My questions are:

1. Is 3 feet too long? If so, what's the appropriate length without causing problems or degradation in power supply quality?

2. Should the 7 wires be braided, twisted or just leave them parallel to each other?

Thanks in advance for your thoughts and advice!

  

3 feet is fine, make sure the wire is 600v rated, I use 18 awg

no filaments so no reason to twist

7th wire should be a chassis to chassis wire, power supply needs to have power input ground tied to chassis

I bought a bunch of different colour coded teflon high voltage rated insulated wire (eBay) when I built the T2 (and earlier than that the original BH). It was actually a bit more complicated with those valved/tubed designs because the heaters floated at 500V (IIRC), so I twisted the heater wires, and then pulled them through glass braid sleeving as an additional insulation barrier to the other umbilical wires. That is not a problem with the carbon because there are no heaters, but making up high quality high voltage umbilicals is time consuming whichever way you look at it..

Thanks Kevin and Craig! 

 

23 hours ago, kevin gilmore said:

3 feet is fine, make sure the wire is 600v rated, I use 18 awg

no filaments so no reason to twist

7th wire should be a chassis to chassis wire, power supply needs to have power input ground tied to chassis

I plan on using exactly that - 18AWG wire, 600V rated, PTFE stranded wire I bought from Apex Jr. 

I will be using a star-schema grounding tying the IEC ground and the regulator ground to the chassis. In this case, should I still tie the 7th wire to the chassis on both sides? 

On the amp side, how should I handle the grounding? Will tying the 7th wire to the chassis be sufficient?  

 

22 hours ago, Craig Sawyers said:

I bought a bunch of different colour coded teflon high voltage rated insulated wire (eBay) when I built the T2 (and earlier than that the original BH). It was actually a bit more complicated with those valved/tubed designs because the heaters floated at 500V (IIRC), so I twisted the heater wires, and then pulled them through glass braid sleeving as an additional insulation barrier to the other umbilical wires. That is not a problem with the carbon because there are no heaters, but making up high quality high voltage umbilicals is time consuming whichever way you look at it..

I did not go into all details - I am actually planning to use the same PS unit for my Blue Hawaii and, in the future, the Grounded Grid. I plan to have a separate 4 pin connector and wires for the 6.3 vac filament supply which I will twist the wires.

I also plan to install a separate switch (together with thermistors) for the transformer secondaries for the HV. This switch will function as a manual delay for the HV supply for the BH and GG. I'll leave this switch in the "ON" position when I use the PS with Carbon and HV. When I use it for the BH and GG, this additional switch will be in "OFF" position when I turn on the main power switch and be flipped on after 30 seconds or longer.   

 

EDIT: I forgot to mention that the reason I am doing the HV delay this way is because I use the off-the-shelf Antek transformer (AS3T-325) which has the 6.3vac and 325vac secondaries on the same transformer. 

Edited October 1, 20169 yr by mwl168

Hmm John, would you be willing to build another Carbon ??

@johnwmclean:

Hi John,

I wanted to ask for a time what for the xlr outs on your Carbon beauty, when do you use them?. They are connected to the inputs, not? I know two connected inputs are usual at Stax but I preferred until now  some switched single ended inputs as second inputs. The problem is the switch, good switch is a hard to find thing, so I am thinking about what to choose.

judo

Hi judo,

The outputs are directly connected to the inputs. The benifit is that I can run multiple amps daisy chained from a single source without having to disconnect and reconnect amplifiers.

John