LuFo Amp - 39w SE Class A from 28v Rail

[xrk971]

There are 2 R121 on the board,is it where the unpopulated R121 the 0 ohms "resistor" should be?

Sorry, that’s a silk screen error on Gen 1 boards. It should be R122 (0R) and use a wire jumper if you are using the aluminum IMS TO247 adapter which has a 100R gate stopper resistor on board already.

 

[Plott]

Many thanks X, I will try to figure it out.
Would the/your SLB work well? And is it still available?

 

[lydlaug]

You can try CRCRC with a high ripple current high ESR 105C rated 15,000uF cap on first C-

You mean low ESR?

 

[xrk971]

No, you don’t want low ESR for the first cap in a CRC as it causes too high of a current surge in the charge cycle which leads to high ripple current. Low ESR is good for last rail cap right next to an amp output to give reserves in a a class AB amp. I should say “average” ESR rather than low.

 

[lydlaug]

Hmmm.
I was thinking that a high (average) ESR would generate more power/heat in the capacitor than a low ESR as the power is
voltage squared divided by R (ESR)...
But yes, on the other hand, the power is also current squared times R...But if the R is lower?
What about a small R ahead of the first C with low ESR?
That way the power generated by the high current surge in the charge cycle will heat up the R instead of the C?

 

[xrk971]

you can run a simulation in LTSpice and see how independent R and ESR affects power dissipation. Those were the results when we ran simulations of the CRC used in cap multiplier for Class A. The specific capacitor used was an important consideration based on results of simulation.

 

[Studley]

X, are those new Lender boards for the LuFo going to be in your store soon?

 

[Ryssen]

Sorry, that’s a silk screen error on Gen 1 boards. It should be R122 (0R) and use a wire jumper if you are using the aluminum IMS TO247 adapter which has a 100R gate stopper resistor on board already.

View attachment 1020633

Ok I thought so,thanks.

 

[xrk971]

X, are those new Lender boards for the LuFo going to be in your store soon?

I have them but they have not been verified yet to work.

 

[Studley]

I’ve emailed Monolith Magnetics and S P Wound to see if either are interested in making some custom chokes at an attractive price.

 

[poseidonsvoice]

@Studley

Thanks for contacting Monolith Magnetics. They are an exceptional transformer company. Hopefully a group buy would get the prices down a little.

Best,
Anand.

 

[Ryssen]

Was it 50vp or 50vpp that was neded to drive the amp fullt?

 

[xrk971]

50.5Vpp for 40w into 8 ohms. So preamp will need a few extra volts since amp is slightly below 0dB gain.

50.5Vpp / 2.83 = 17.85 Vrms

17.85^2/8ohms = 40wrms.

 

[Ryssen]

Is there a place for V122 and V162 on the PCB?
I cant find it..

 

[Ryssen]

Still a long way from finishen,but I still wanna know hom to adjust the 2 trimpots.

 

[xrk971]

One sets the bias current and the other sets the gate voltage for the JFET. The two are linked so it requires iteration. I have not personally tried myself as my boards did not have the second pot. I think Mbrennwa used it to set exactly the current and voltage independently.

 

[Ryssen]

Ok what valde (voltage) should they be set at?

 

[xrk971]

I would try to get setpoints to match what is in the simulation here:

 

[von Ah]

What’s the consensus on appropriate chokes in 2022?

I know there has been success with MOT MD-803s btw xrk and Vunce.

Xrk, you also tested a MD-103.

dbel and jwjarch tested a MD-903.

Hammond has 195T5 and 195S4 as assumed good options.

Found this on Mouser: https://www.mouser.com/ProductDetail/Bel-Signal-Transformer/CH-4?qs=0/2GmT7MRM9RsVUzZKxOJw==
Would that work, or is it the wrong kind of inductor?

I think the 195T5 would be the absolute highest cost I would go for.

 

[xrk971]

For the price, it’s tough to beat MD-803. About $30 with shipping.

https://www.ebay.com/itm/Microwave-...2349624.m46890.l49286&mkrid=711-127632-2357-0

https://www.ebay.com/itm/TRANSFORME...2349624.m46890.l49286&mkrid=711-127632-2357-0


.