F5 Cascoded - an alternative for my interest in the Balanced F5

Status
Not open for further replies.
Jump to Latest
#83
Finally up - Biasing

My move took a lot more effort than I expected - apologies for the delay reporting back on the amp. I finally unpacked the Cascoded F5.

I have limited toroids, the only one I have besides the 18V is a 156VA 26V output from a set of Logitech computer speakers that lost it's Pod.

I hooked up the 26V toroid to a rectifier and CRC filter and got +-36V rails - not too bad. I biased up one channel of the Cascoded F5 to see what I could get. I'm currently at 1.05 amps per side of the circuit - 2.1 amps total (only one output per side). My IRFP (9)240 case temps are 80C - which seems too high, but I'm not sure. My heat sink temp is 47C with ambient at 24C.

For now, I'm happy. Tomorrow I'll try some music, then perhaps an other output to get my biasing up more.

The only difference I've noticed at the moment compared to a stock F5, is the Cascoded F5 seems to take longer to stabilize.
 
#87
I used an all copper washer. I'll post a pic tonight.

I think its the change in energy dissipation. With a stock F5, one dissipates 31W, with the Cascoded F5, I'm dissipating ~40W. I think a second output will solve the issue.

Luckily, I have matched outputs waiting in the wings.


Tea-Bag said:
80C is too high for case temps. Try using a shoulder washer before the bolt into sink, it may help it get better contact.

Hey it works, which is good.
Click to expand...
 
#89
ZM,

I'm going to add parallel output devices. Cviller's (attached) has a 1K resistor (R117 and R118) on his cascode schematic that you do not have in yours. It there a correct way to add more outputs?


Zen Mod said:
there are at least 4 cascoded F5 schematic posted , all of them pretty much similar .....

dunno - I think that AR2 even tried this one

I didn't , but I can't see why it shouldn't work :rofl:

( full blast voltage version really needs few values/parts endurance confirmation)
Click to expand...
 

Attachments

  • Cascoded F5.png
    Cascoded F5.png
    67.7 KB · Views: 330
#94
Tea-Bag said:
I recommend Mica one's off ebay!
TO-247 Transistor Mica Insulator,Insulation sheet, x 50 - eBay (item 250608863278 end time Feb-27-11 00:00:43 PST)
Click to expand...

I bought a batch of those, but I understand they are 4 mills and need to be split to 1 mil. I'll have to buy some sort of measurment device, micrometer or something to make sure the thickness is right. How do you come to the recommended 1 mil thickness? I understand they can be easily split with a razor blade, but how do you measure the thickness your self? I figured I could just split them in half, then split the halves in half. I dont know how consistent they are in thickness, or how critical the 1 mill measurement is.

Thanks,

Russellc
 
Last edited:
#96
ZM and TB,

You have both been an invaluable resource on the forums. Without spending more than a second, can you elaborate on the mica vs Keratherm issue?

I can see convenience with Keratherm, but it always takes weeks for my shipments to come from Zhoufang. Ebay is often fast (for me). Mica seems straight forward, but I can see the hassle factor of splitting the pieces.
 
#98
NYCOne said:
ZM and TB,

....Without spending more than a second, can you elaborate on the mica vs Keratherm issue?

I can see convenience with Keratherm, but it always takes weeks for my shipments to come from Zhoufang. Ebay is often fast (for me). Mica seems straight forward, but I can see the hassle factor of splitting the pieces.
Click to expand...

as EUVL sez - that's old story , proven more than once .

in any case - better late than sorry - spend few greenies and buy keratherm thingies for next few years

its nice knowing that you use best you can , and that without need to sell left kidney

making amp must not be competition in speed , but enjoyable process which will help you to feel Buda ( put there name of your favorite autority or deity , even if your wife is that :rofl: )

:clown:
 
#100
Parallel Outputs...getting there

I set up for parallel outputs on my heat-sink tonight. I haven't had a chance to try it out. Tomorrow I'll see where I can get to with bias. I'm ordering new insulators, but they'll take time to come.

I could use advice on a few fronts:

  1. Assuming I use the right voltage dividers, how high will the cascoded F5 reasonably go with rail voltage? I've asked before, but never really got an answer I understood. At the moment, I'm limited by my CRC filter. The caps are rated to 50V.
  2. If power output is calculated by P=I*2R, what's the advantage to higher rail voltage? Isn't the advantage for power found with a higher bias current?
Finally, is there any reason for me to replace the BC546 and BC556 with low noise versions (BC550 and BC560)?
 

Attachments

  • Outputs.jpg
    Outputs.jpg
    139.5 KB · Views: 279
Status
Not open for further replies.
Top Bottom