Board for WA1GFZ MOSFET Audio Driver

[w9jsw]

Here are all the docs. You can upload the xls to digikey and use the quantity and the supplier part number to build your cart. I tested it. I did not try at mouser but you should be able to use the mfg part number. You may have to sub parts if they don't carry that exact part. I used the digikey parts to make sure the footprints were adequate. If you sub parts, you are on your own with respect to them fitting ok.

Mount the large 10W resistors slightly off the board. Also mount R14/R16 slightly off the board.

You will need to order the bias pots separate. They were crazy expensive at digikey. You can buy good pots at the surplus dealers for much less.

Good Building!

John

[W1RKW]

John,
Received my board just now.  Looks good!
Thank you again.
Bob

[w9jsw]

Just bought a pair of these for the bias pots...

https://www.ebay.com/itm/Clarostat-140-6130-100K-Ohm-Pot-Potentiometer-5905-174-4794-2-Watt-New-Honeywell/252809317229

John

[W1RKW]

Updated the BOM for sourcing the caps and pots via Newark.

Priced out the unit cost for Digikey and Newark.

[KB2WIG]

J,

Rcvd da boards, FB OM.  They are as pretty as a fresh water clam.... ..

klc

[w9jsw]

Build pics from the maestro himself.

His needs are modest - using it in a Valiant, so not a big heatsink. For a larger tube driver, we suggest 1in fins.

Note how he drilled and tapped the sink to mount the FETs. Used 3/8 washers but 1/4 may also work if you keep your thru hole leads trimmed well.

[w9jsw]

Updates on fusing and bias

[W1RKW]

with 5 FETs sharing a heatsink, is there any concern for one or more FETs heating up and changing bias and idling currents of other FETs on the same heatsink?

[K1JJ]

with 5 FETs sharing a heatsink, is there any concern for one or more FETs heating up and changing bias and idling currents of other FETs on the same heatsink?

Hi Bob,

I suppose anything could happen and produce a thermal runaway situation if taken to an extreme.

But in normal operation all the MOSFETS run pretty lukewarm and are very stable with capacitive coupling between stages.  Even if one FET were to create a problem I think the normal power supply fusing or the grid fusing might protect it from an extended heat up. But ya never know when it comes to voltage spikes, oscillations or other dangers to SS devices.

I have always been an extremist when it comes to circuit protection. My first class E rig before circuit boards were available had about 7 different sensors and shutdowns. It was foolproof even sensing input and output swr.  Also had a diode on the heatsink for temp sensing, etc.  Steve went on to build a nice all-inclusive protection design with Hall effect devices that did it all and more -  and is still available today.

So in the case of the GFZ driver board, there's no reason why you couldn't add a temperature probe on the heat sink to sound an alarm above a certain temp threshold. I have them sampling warm air above the finals in all of my bigs rigs. I use a digital/LED Kiln temp controller module that can be easily programed.

These are the type I use for $29.50. It would be easy to implement in the driver board. The top readout is the actual measured temp and the bottom is the preset trigger alarm limit. Easy as pie to use.  Comes with a sensor. (Our Chinese friends)

https://www.ebay.com/itm/12V-24V-32V-PID-SSR-TEMPERATURE-CONTROLLER-FURNACE-KILN/230399708125?epid=1448454664&hash=item35a4e48bdd:g:L7AAAMXQbcRQ6RbM


T

[w8khk]

Another factor that determines thermal stability is the circuit topology.  If several stages are direct-coupled, (Direct Current) then a change in bias in the driving stage will likely cause a more dramatic bias change in the driven stage.

The board we are looking at here has resistive voltage dividers setting the bias for each individual stage, and the output push-pull stage has the bias individually set for each device.  All stages are capacitively coupled, passing only the AC audio signal, not the DC bias.  This results in a very stable design, and therefore thermal runaway due to the common heat sink should not be a concern.

The same device types are used for the amplifier, phase splitter, and the output stages.  The lower level input stages run at much less current with higher resistances than the output stages, further contributing to the reliability due to their conservative operation.  Tom's success with the design for a decade speaks volumes!

[w9jsw]

Build is progressing at Frank's place...

[w8khk]

My second board arrived this afternoon.  Beautiful work...  Thanks, John!

[KD6VXI]

I've been away from the house the last two weeks....  But, got home and my board is there.

Thanks a million.  One of these days I might get some time to out it together and transplant into the Viking!

--Shane
KD6VXI

[w9jsw]

Updated BOM.

I had the wrong fuse clips. Now corrected.

Integrated the part numbers for newark for the caps and pots.

Anyone other than Frank building?

John

[W1RKW]

 Not at this time.  Trying to wrap up my crystal radio project then with warmer weather around the corner will be outside.  probably won't start collecting parts until summer time.

[w9jsw]

Me too,

Taxes done, finally,

Helping kids remodel a bath,

2 bedrooms to replace carpet with wood floor,

Going to be a while...

John

[KD6VXI]

I wish.

Haven't seen home in almost 6 weeks, sans a day every week to do laundry and check that the house hasn't burned down.

Got into the IBEW, new job 150 miles away.  Finally out of Bakersfield and back to the 'coast of Cal'.

Signed papers on the new h ouse Monday.

Gonna be a couple more weeks, lol!

--Shane
KD6VXI

[Opcom]

That board sure has turned out beautifully. It's as tasteful as the table upon which it is shown.

[KB2WIG]

After I bite the bullet on the pot$, and waddle through the 4.2K listings of heat sinks on Mouser, then I'll start.... after the spring antenner work.

klc

[K6IC]

Updated BOM.

   ...   Anyone other than Frank building?
John

Hi John,

Spring has sprung here,   so  were back to trying to abate Fire Risk from weeds,   tree debris,   and general sloth  --  lotta rain this year,   so the WEEDs are getting ahead of us.

But hope to try to get to building a board,   this Summer.

Thank you again,   and to all others who have contributed to this project.

73,  Thanks,   Vic

[w9jsw]

Ordered parts for my board. Right at $100 from Digi including shipping and taxes. Did not order the pots (got on the 'bay) and the caps. Plan to place the supply on the modulator chassis right next to the board so the main caps on the supply should be enough. Going to use an Antek transformer for the supply and build it per the design in the design notes doc. 

Modulator assy will be a 17 wide by 22 or 24 long piece of 3/4in plywood base with a 10x17 aluminum chassis for the mosfet board and the 813's. Hoping a 17x12-14 area is big enough for the mod trans, mod reactor and cap. That Gates iron is big and heavy. If not big enough, I will make it longer. 

John

[w9jsw]

Trying a few parts for fit. Was some discussion off-line on fuse clips. Here are the proper clips as spec'ed in the BOM and a 5x20 100ma fuse. Fits just right. Also note the coupling cap in the background. It fits just right also...

Hopefully I can get some more parts placed this weekend. Going to be rainy so there is hope!

Wondering if my 15W pencil soldering Iron will have enough gas to solder this board. I bought it for SMT soldering, not this stuff. My alternative is a Weller Gun. What do you guys use? Perhaps a 30-50W pencil? Suggestions please...

John

[steve_qix]

Can't speak to wattage directly, but I use an adjustable soldering station.  Usually for components of this size, I run it quite high in temp (as high as it goes, usually).

[W1RKW]

Ordered the parts for my board.

found a suitable heatsink with a reasonable price.  Seller has a few more available.

https://www.ebay.com/itm/Heat-Sink-938SP-02000-A-200/121410536527?hash=item1c44a1c44f:g:0RwAAOSw7NNT7PBG:sc:USPSPriorityFlatRateBox!06415!US!-1

[w9jsw]

Nice Heatsink Bob. Looks like it was almost cut for your board. 6 more avail it appears. Inexpensive as well.

I picked up a 65W pencil iron on Amazon today. First time I have seen a product there with no negative comments. Hopefully it will work out. If not, was not that much $$.

I am going to drill and tap the heatsink before I start populating parts.

Strongly suggest to not skip the outboard filter caps. (C5/C6). You can fiddle with the values but we are seeing that they matter if your supply is away from the board. I plan to go with at least 470uf and my supply will be on the same chassis. The board has such a good audio passband that even minor 120Hz ripple is audible if not careful on filtering. This board provides really crisp/clean audio. We may have an errata on grounding as testing proceeds.

John