Group Build: Dynafet

[Torpedo]

Definitely, if it weren't for my friendship with PICaudio (Rodrigo) I could only participate in the non-Diyer version which just required soldering a few connectors and screwing boards in place. But I think this was the original idea, that people not being skilled DIYers themselves could get the tough work done by someone else, while the pure DIYers could get every single bit and piece apart to build the whole thing by themselves.

[Smeggy]

I'm still hoping that a pair of populated boards is all I'll need along with a little help with info on strapping the rest together.

[n_maher]

Well my offer to stuff boards is certainly still on the table, just let me know if it's needed.

[DigiPete]

Well my offer to stuff boards is certainly still on the table, just let me know if it's needed.

As is mine. If the enclosures are predrilled I could contemplate assembling an amp or two (but that remains to be seen).

[PICaudio]

Hi guys

Since my friend Torpedo spoke to me of the existence of this group, I have followed with great interest everything that happens here. I have always liked the designs of Kevin and I see that you are working hard on a proyect with a high-level desing.

There are people with very good ideas in this group, but I think we should go step by step for doesn't confuse the non-DIY members.

I think we should know the approximate price of a basic configuration of Dynafet with 2 ch and 4 ch to determine how many units are to be built, if a user wants more quality components, or a more complicated desing (microcontroller, I2C-bus, input transformers, etc)... he already know that have to pay more.

Like we live in Spain (Torpedo, Hitoridekimasu, and me), it's likely to be interesting to us buy directly the "heavy" materials (boxes, toroidal transformers), because the dispatch from US may be more expensive than the money that we can save buying with the group. We will not have any problem to make the machining of the boxes.

I don't have many problems to understand your comments in english, but for me it's difficult to write in this language , and the google "translator/language destroyer tool" doesn't help much :)

-- o --

Rodrigo

[DigiPete]

Hi guys

Since my friend Torpedo spoke to me of the existence of this group, I have followed with great interest everything that happens here. I have always liked the designs of Kevin and I see that you are working hard on a proyect with a high-level desing.

There are people with very good ideas in this group, but I think we should go step by step for doesn't confuse the non-DIY members.

I think we should know the approximate price of a basic configuration of Dynafet with 2 ch and 4 ch to determine how many units are to be built, if a user wants more quality components, or a more complicated desing (microcontroller, I2C-bus, input transformers, etc)... he already know that have to pay more.

Like we live in Spain (Torpedo, Hitoridekimasu, and me), it's likely to be interesting to us buy directly the "heavy" materials (boxes, toroidal transformers), because the dispatch from US may be more expensive than the money that we can save buying with the group. We will not have any problem to make the machining of the boxes.

I don't have many problems to understand your comments in english, but for me it's difficult to write in this language , and the google "translator/language destroyer tool" doesn't help much :)

-- o --

Rodrigo

Rodrigo,

Welcome to Head-Case!!

Cheers,

Pete

[grawk]

You're better at english than I am at spanish

[Dreadhead]

I'm in the boat that populated boards or fully built are fine with me. Vlad I and I will work something out wrt the overall build.

If we are going to go with professionally populated boards I may buy some extras for backup etc.

[Icarium]

Okay here's what I propose. We keep the group buy for now to boards,

board parts, sigma22 boards, sigma22 parts, SumR toroids, Sowter

transformers. Chassis, other boards, volume control and other parts

will come later.

I will at the very least organize the toroids and the sigma22 board

buys (amb is local to me so I can just pick them up and mail them). If

I can float it then I would be willing to kit out the sigma22s as well

by picking up parts (If this requires matching then... yeah nvm I can't

do that :/). If I can float it I can cover the dynafet boards too.

Luvdunhill will be in charge of organizing the dynafet transistor effort.

After we figure out who is buying what I will try and figure out a money exchange

scheme that makes the most sense. I think it'd be best if we distribute money to the

procurers after all procuring is paid for to figure out the path of least fees, but that is

highly dependent on if procurers can afford the initial hit and if they feel comfortable with

doing so. I personally am as long as I can afford it. I am also willing to accept

money order or personal check if the payer wants to minimize fees.

So if people feel comfortable with everything I propose then I will start a new

thread to figure out who exactly is opted into what. Hopefully we can lock

in the people who are still interested within a week and begin ordering things

that it makes sense to order.

Please give me any feedback ASAP and tomorrow I will begin a new thread with an

updated list of what people seem to be interested in and if people could confirm that

they are still interested in that thread. Please really think if you can see this through as I

really do not want to eat the cost of stuff if people back out.

For non-DIYers who do not have someone to do the full build for them

our esteemed board layout meister j4cbo has offered to do some full builds

in order to subsidize his own build. There are still 1-2 spots available. Please PM

me if you are interested and we will figure it out. There will be more spots if

people are interested in single ended versions (Less labor). I need to have an

answer within a week or whenever we lock down who is interested and begin

making purchases.

This is the general design of what the end result will look like. The

result will be something like this:

- Volume control will be a resistor-ladder based attenuator, similar

to the Joshua Tree, of his own design as well (to keep board count

low).

- Sowter transformers as previously mentioned for unbalanced to

balanced input and vice versa (2 are required, left and right).

- Front panel user interface will be with a rotary knob and VFD

display; this will allow input and output selection, thermal and power

monitoring, and volume control without cluttering up the front panel.

-Two chassis or single chassis.

-Number and type of outputs... flexible.

-Number of inputs is also flexible.

-We can discuss other features, but some may get veto'd due to difficulty and

anything extra fancy will cost extra.

He estimates that cost without chassis will be between 1050 - 1350 for a balanced

dynafet. Let's just round that up to 1500 for unexpected expenses. If two of us

take him up on his offer and we completely subsidize his cost we'd be looking at paying

2250 each. If 3 of us enter this then we are looking at 2000. This is without chassis.

He may have more slots if people are more interested in single ended versions. Parts

cost for that without chassis will be ~600 dollars and amount you are subsidizing

will drop accordingly.

Now chassis options are this we can either go with something pre

fabbed like this: modushop.biz

and having my brother machine any holes we need. He is willing do to

this very cheaply (Possibly for some subsidized parts for himself),

But a 1 month membership to the machine shop s $125 dollars.

TechShop is the SF Bay Area's Only Open-Access Public Workshop -- Start Your TechShop Membership Today

This cost will be distributed as well.

If we decide to go this route and DIYers are interested as well then

he says he could probably easily process all the chassis we need in

the space of a weekend. The only tricky thing is we might need to

order some extras just in case something happens. Like probably one to

test on and possibly a couple extra in case some get damaged in route

or they get damaged in our possession (I don't know, but given that

company is out of the country... ordering some backups just seems to

make sense).

The other option is to design something completely custom and he would

be willing to machine this as well. But if we do this then cost could

be pretty expensive. It'd be 150 for raw materials if we go with a

thicker plate chassis... less if we go with sheet aluminum. Then he'd

want something for his time as well and the limit would probably be ~5

chassis as it would take quite a bit of time to machine this. We could

also anodize which he estimates to be 200 for a batch (Distributed across

2-3 people at the least).

So really it's what you want to spend for chassis but expect to spend

~150 at a minimum for a single chassis and 300-400 if you want two. If

we go custom then he estimates 300-400 per chassis if we go with some

thing simple. Probably closer to 300.

So we are looking at basically 2500 - 3000 for a pretty pimped out build. It may

cost less or it may cost more. No guarantees at this point in time... we will be

very transparent with where money is going and why it is needed. If you want to

skimp to save money then that is perfectly workable.

Timeframe is this. j4cbo is pretty committed to getting everything done by the

end of 2009, but there can be no guarantees. This may push well into 2010. He

is coming to Mountain View during the summer for an internship which is where

my brother and I live so depending on our free time we can push this project

pretty hard during the summer. Summer completion is.. possible, but do not

count on it.

What we'd be looking for as partners in this.. people who can afford it and are

absolutely committed to not flaking and seeing it through. There can't really be

any refunds.

The end result should be pretty glorious though.

[Icarium]

Input selector schematic.... one half of his uber micro-controller board.

[Torpedo]

How much raises cost having the controller board for monitoring temperature, selecting source and controlling the volume stepped attenuator compared to mounting a good volume pot and analogue selector switch?

Not very worried about the raw cost, but for the noise and additional circuitry and complication it means.

[n_maher]

Just got this form Hi-Fi 2000, figured it might interest builders.

We carry to acquaintance our Dear customers that we have put on-linens a new Pesante DISSIPANTE 5Unita deep 500mm for the realization of amplifiers of large power. modushop.biz We signal you that we have perfected our service of perforation and milling , having offered also the service of engraving of the small words, thus allowing the realization of prototypes complete to a contained cost, without the costs of expenses system to the system of press with serigraphy. on ours link it is possible to consult the complete list of the workings. And the prices modushop.biz Hifi2000

[Icarium]

How much raises cost having the controller board for monitoring temperature, selecting source and controlling the volume stepped attenuator compared to mounting a good volume pot and analogue selector switch?

Not very worried about the raw cost, but for the noise and additional circuitry and complication it means.

Noise I don't know.. maybe j4cbo could comment. Complexity-wise it'll just be another board, but will require smd soldering. I think our builds are going to be 4 dynafet boards, a micro controller board, 2 sigma22 boards.

Ah cool Nate! Thanks for hooking that up. Yeah I'm guessing we'll go the pre fabbed chassis route.. custom just seems too high, but eh we have time to figure that out.

[j4cbo]

How much raises cost having the controller board for monitoring temperature, selecting source and controlling the volume stepped attenuator compared to mounting a good volume pot and analogue selector switch?

Not very worried about the raw cost, but for the noise and additional circuitry and complication it means.

I think the controller board compares favorably with stepped attenuators. A 4-channel Goldpoint attenuator plus a 2-deck selector switch winds up costing $350, and I'm estimating something like $250-$300 for the controller board in a balanced-only configuration, including a graphical VFD and knob. It'll be more like $150-200 for a two-amplifier-board configuration. Those prices don't include the additional $130 for a pair of transformers, if converting between single ended and balanced is desired.

Electrical noise will be minimal. It's possible to shut down the microcontroller's clock entirely and turn off the VFD when the control interface isn't being used, but I don't anticipate that being necessary; I'll be careful to design the board for good separation between digital stuff and the actual signal path, and the UI portion can be shielded off from the rest of the amplifier.

[Torpedo]

Thanks for the comment j4cbo. If the controller board compares that favorably price wise and includes the resistors for the volume control, it makes sense. However dealing with smd soldering can be a problem for some. Maybe leaving that optional if possible might be interesting.

[j4cbo]

The controller board is designed primarily for the complete amps that Icarium, myself, and whoever else is interested will be building. For those, I'll be populating them myself as well as writing the firmware.

That said, if other people are interested in using the design, I wouldn't mind selling the boards and preprogrammed processor chips. I could also just solder the surface mount portions, and leave the rest to be populated by the buyer.

[Torpedo]

Very good option. I'll wait for Picaudio's comments since he'll be building my amp. Maybe the spanish group sums to it.

[Icarium]

http://www.head-case.org/forums/casino/5732-group-buy-dynafet-contd.html#post208187

Okay as promised here is the thread specifically tailored for group buy discussion.

The reason I separated it out to a different thread was to avoid it being mired on page 23/24.

Please read it over and check the spreadsheet. If your name is not in Bold or in the appropriate section please let me know asap.

Bold means locked in and ready to commit to the initial dynafet transistor buy.

Out means that you've pulled out.

Thanks!

Btw some of you might notice the second spreadsheet. I have created that for organizing who is populating who's boards. If people are comfortable with it then I'd like to put at the very least what state people live in so we can coordinate shipments and board population in a way that makes geographical sense. If you guys are extra comfortable with zip codes then I can also use that to estimate shipping costs accurately when it comes time for that without relying on an exchange between me and you.

What would be awesome is if some of you (Like Dreadhead/vvs_75 or Team Spain though they won't really be interested in group buys past boards/transistor) are close enough to receive a single package for multiple parties. That will definitely cut down on confusion/shipping costs/packing pain.

[Smeggy]

Is there any closely estimated cost per populated board yet?

[Torpedo]

Yep, I can receive the boards and parts for the Spanish team and deliver them to Hitoridekimasu and Picaudio, or maybe it's more sensible Picaudio receives them (since he's building my boards) and just sends to Hitori his share. No worries, we'll try to make things easier for you

Also if you are sourcing parts from Europe which we are using too, we could order and get them, then sending the US part to you, perhaps that saves you shipping costs or custom duties. Even if we weren't using any of them, if some of you needs parts from this side of the pond and we can be of any help, just let us know.

[luvdunhill]

Okay here's what I propose. We keep the group buy for now to boards,

board parts, sigma22 boards, sigma22 parts, SumR toroids, Sowter

transformers. Chassis, other boards, volume control and other parts

will come later.

Agreed. I think this the best way to go forward. I'd like to handle the Sowter group buy as well, as I need 4 input transformers just for myself. I have worked out a custom version from them that should be amenable to everyone. I'm also planning on offering this to builders of another project at DiyA as well.

I will at the very least organize the toroids and the sigma22 board

buys (amb is local to me so I can just pick them up and mail them). If

I can float it then I would be willing to kit out the sigma22s as well

by picking up parts (If this requires matching then... yeah nvm I can't

do that :/). If I can float it I can cover the dynafet boards too.

AMB will sell matched transistors for the sigma22 boards. You may also consider working a deal out with Jeff at glassjaraudio for some discounts. I think this route would be easiest, as there are a lot of parts to kit.

but that is highly dependent on if procurers can afford the initial hit and if they feel comfortable with

There is pretty much no way I can do this for everything. I can and have for the Toshiba JFETs, but that's about all I can do. I'll have to wait for money before I can order the small heat sinks and the small signal BJTs and the output devices.

For non-DIYers who do not have someone to do the full build for them

our esteemed board layout meister j4cbo has offered to do some full builds

in order to subsidize his own build. There are still 1-2 spots available. Please PM

me if you are interested and we will figure it out. There will be more spots if

people are interested in single ended versions (Less labor). I need to have an

answer within a week or whenever we lock down who is interested and begin

making purchases.

I can also do a build for someone as well. I'd prefer a balanced build.

He estimates that cost without chassis will be between 1050 - 1350 for a balanced

dynafet. Let's just round that up to 1500 for unexpected expenses. If two of us

take him up on his offer and we completely subsidize his cost we'd be looking at paying

2250 each. If 3 of us enter this then we are looking at 2000. This is without chassis.

That sounds reasonable.

Now chassis options are this we can either go with something pre

fabbed like this: modushop.biz

and having my brother machine any holes we need. He is willing do to

this very cheaply (Possibly for some subsidized parts for himself),

But a 1 month membership to the machine shop s $125 dollars.

TechShop is the SF Bay Area's Only Open-Access Public Workshop -- Start Your TechShop Membership Today

sounds like a good way to cut costs! One other option is to order heat sinks from Conrad that have already been milled and tapped. Then you're just looking at the top and bottom, which can be cut easily and the front and rear custom panels. In this case, things are very easy to assemble and will be cost effective as all that is shipped from overseas is the heat sinks. I've worked with R-Theta as well, as since they are closer, they may offer us a better deal. In the case we're interested in Conrad, I have a contact there that has done this exact thing in the past and we may be able to use previous drawings. The final product, including shipping in the US to the final destination was under $100 a pair. Spritzer has a pair of these heat sinks, they are extremely nice. Just another option, one I think would be the cheapest, assuming your brother can cut the top and the base plate and do the machining on the angle brackets.

So really it's what you want to spend for chassis but expect to spend

~150 at a minimum for a single chassis and 300-400 if you want two. If

we go custom then he estimates 300-400 per chassis if we go with some

thing simple. Probably closer to 300.

$150 is low. I'd say $300 is spot on.

Thanks for your work guys, sorry I was AWOL... I was out fighting crime and all that stuff...

[luvdunhill]

don't forget that the input transformer likes to see a certain load after it. Please add a resistor in series with a capacitor after the input transformer so the loading can be adjusted.

[luvdunhill]

I think the controller board compares favorably with stepped attenuators. A 4-channel Goldpoint attenuator plus a 2-deck selector switch winds up costing $350, and I'm estimating something like $250-$300 for the controller board in a balanced-only configuration, including a graphical VFD and knob. It'll be more like $150-200 for a two-amplifier-board configuration. Those prices don't include the additional $130 for a pair of transformers, if converting between single ended and balanced is desired.

Electrical noise will be minimal. It's possible to shut down the microcontroller's clock entirely and turn off the VFD when the control interface isn't being used, but I don't anticipate that being necessary; I'll be careful to design the board for good separation between digital stuff and the actual signal path, and the UI portion can be shielded off from the rest of the amplifier.

Will you use another transformer (PCB mount?), or perhaps an external wall-wart to power this circuitry? Sounds good, but I wouldn't really want to tap the main toroid for this.

[luvdunhill]

with regards to Pars post in the business thread, I'd definitely go with the larger transformer and go with the same transformer for all builds. I'd recommend 160VA 2x30 VAC with universal primaries. Again, reference Plitron 057017201.

With regard to SumR, they are a bit of an unknown. The ones I have seem to have a bit of mechanical noise, but perform okay. Plitron is pretty much the industry standard in audio. I just wanted to throw this out to the crowd, and maybe those who have an opinion on this matter can give their $0.02. This is pretty much a no-compromise type of build and from what I've seen so far, this is the only already I'd consider a slight compromise, just because of the unknown-factor.

... gotta run for a bit, I'll comment more on the other thread later..

[naamanf]

With regard to SumR, they are a bit of an unknown. The ones I have seem to have a bit of mechanical noise, but perform okay.

If this is really the case and we are going with a one case design (or even a two for that matter.) I think we would be better off spending a little more for something that isn't going to cause problems down the road.