"Blameless" style ~100W amp with CFP output. - Page 2 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Thread Tools Search this Thread
Old 1st April 2011, 07:34 PM   #11
diyAudio Member
Join Date: May 2010
Yeah, I understand that. They're rated at 500mW and they're dissipating less than 250mW. That doesn't worry me too much. The don't seem to be getting too warm, but if they do I'll replace them.
  Reply With Quote
Old 1st April 2011, 11:02 PM   #12
diyAudio Member
5th element's Avatar
Join Date: Oct 2002
Location: England
Originally Posted by mnturner View Post
Yeah, I understand that. They're rated at 500mW and they're dissipating less than 250mW. That doesn't worry me too much. The don't seem to be getting too warm, but if they do I'll replace them.
That isn't what's being said here. The transistor can cope with 500mW when @ or below 25 degrees C.

The case style will increase in temperature by 250 degrees per watt.

Given 17 volt rails at a CC of 6.7 mA you are indeed looking at 0.11mW, which will result in the transistor running at around 45 degrees, which is not worryingly hot to the touch.

However if you increase the rails up to 30 volts, necessary for 50+ watts @ 8 ohms and you're looking at around 200mW. This will increase the transistors temperature by 50 degrees, add in an ambient of 25 degrees and we're up to a 75 degree working temperature.

These transistor types can typically handle around half their rated power @ ~75 degrees.

This will probably be fine run outside of a case, where it will experience pretty decent convection cooling, but put inside a case, where the ambient temperature will be higher and you've got a potential situation on your hands.

Originally Posted by mnturner View Post
But if they do I'll replace them.
This is the wrong attitude as it implies you're already running them under the circumstances that would necessitate a change, ergo they could explode before you realise they are running too hot. Or you could burn yourself when seeing if they are too hot.

We've all been there at least once before anyway, just be warned that the VAS and CCS transistors will require heat sinks if any decent reliability is to be expected, when run at higher voltage rails and when inside a case.
What the hell are you screamin' for? Every five minutes there's a bomb or somethin'! I'm leavin! bzzzz! Now with website! www.5een.co.uk under construction.
  Reply With Quote
Old 2nd April 2011, 06:30 AM   #13
Mooly is offline Mooly  United Kingdom
diyAudio Moderator
Mooly's Avatar
Join Date: Sep 2007
Remember the Vce across the transistors isn't constant too... think what happens as the output swings.
  Reply With Quote
Old 2nd April 2011, 08:25 AM   #14
diyAudio Member
Join Date: Nov 2006
Location: Adelaide, South Australia
Blog Entries: 45
I think that the choices you have made for devices is fine for testing and prototyping. there is noting like getting your hands dirty to get your head around things and learn some tricks.

That said, in a power amp of more than a few tens of watts, I would use a somewhat more robust transistor for the Vas and current source as discussed above.

In Australia your local shops (Jaycar, Altronics etc) just don't stock good devices for use in Vas (T8). I suggest that you look on EBAY for a good Vas transistor. I have very successfully used 2SC2682 - look on other forums for a wide range of other options.

For the PNP current source, (T10) I suggest you consider the MJE350 which works fine and is easily available.

With this change the issue of power dissipation goes away - as does the upcoming issue of voltage rating which is yet to be raised.

Why voltage rating? for 50-60Watts, you will be looking at rails of > +/-40 Volts. Which will result in peaks of 80 Volts on T8 and T10. Not a happy place to put a BC556 in!

A question though - what led to the decision to use TIP2955 / 3055 as output devices?

If you used the commonly available MJL21193/4 your would get better Hfe, easier loading on the Vas stage (perhaps allowing you to drop the standing current and ths dissipation) and possibly less distortion. These devices ARE available easily at your local electronics shops in Australia. By the way, the MJL21193/4 are very - very rugged in the deal.
  Reply With Quote
Old 3rd April 2011, 09:18 AM   #15
diyAudio Member
Join Date: May 2010
For the moment the amp is sitting on the bench. which could indeed by why I haven't had any power dissipation problems with the VAS transistor. Once I find something suitable, I'll replace it. What I'm looking for is something with high Hfe, >80v voltage rating, low Cob and decent power handling yes?
I used TIP2955/3055's because (at jaycar at least) I could get 4 pairs of them for the price of one pair of MJL21193/4 and wasn't totally confident the thing wouldn't just blow up as soon as I powered it up.
I'm thinking this one might become a guitar amp where the slightly increased distortion won't bother me, but I like this amp and will probably build a couple more for proper stereo service in which case I'll definitely use MJL21193/4.
Thanks for the feedback guys.

Googlyone, wouldn't i get 60W out of roughly +/-32V rails? I thought +/- 40V would give close to 100W of sine power?

Last edited by mnturner; 3rd April 2011 at 09:28 AM.
  Reply With Quote
Old 3rd April 2011, 09:58 AM   #16
diyAudio Member
Join Date: Nov 2006
Location: Adelaide, South Australia
Blog Entries: 45
If your amplifier can put 40 volts peak across an 8 ohm load, you will get 100W sine power. The problem in practise there are inefficiencies and losses - and from the perspective of an amplifier the ratings of devices need to consider unloaded and loaded conditions.

What does this mean? I won't try posting a whole amplifier model (I have a heap of spreadsheet type analysis I use in designs - but they make MY eyes water...) A few things you need to worry about are:

To get 100W into 8 ohms, you are right, you need 40V pk.
- At this point you have 5 Amp peaks
- Your power supply will typically be a transformer / bridge / capacitor arrangement. Assuming a decent toroid, big bridge and modest capacitor bank (say 10,000uf for a 100W amp) you will see rail droop in the order of 6 or 7 volts from the unloaded condition. (seems like a lot? it is!)
--> From loaded to unloaded supply you need to budget 7 volts droop. Larger capacitors help, as will a larger transformer - but reality is this is life.

- Your amplifier cannot get the output to the supply rail. In your design the use of CFP output helps, but there is the best part of 1 volt drop on the "VBE" of the compound output pair, plus the voltage drop across the emitter resistor, 0.5 volts for 5A pk. Then you have the Vas stage on the bottom half, and current source load for the top half. This can be driven to saturation - and can in fact get to closer than you might think to the supply rail, but is still the best part of 0.6 Volts (have not looked too closely here).
- So you wind up with the Amplifier being able to drive to within a couple of volts of the supply rail under load.

Where are we at:
- Supply rail droop 6 or 7 volts
- Amplifier capacity to drive to rails 2 volts

A 100 Wat amplifier needs 40V pk on 8 Ohms, so is looking for a supply rail of +/-50 Volts. huh? Yep that is the way it works.

Not sure of your age - I think you are Australian though - If you recall the Series 5000, used 50 volt rails for a 100 Watt amplifier. There were plenty of 50 and 60 watt amplifiers (ETI480, 477, silicon chip copy of 480) these used +/- 40 volt rails for a 50-60 watt amplifier.

Crazy big supply capacitors is one way to address this in part, but is expensive and ultimately causes other problems. Simply upping the unloaded supply rails and considering power supply loses in the design is the only sensible approach. The design of an "amplifier" is thus more than the circuit of the power amp itself, you need to look to the whole chain.

Happy to discuss further if you want.
  Reply With Quote
Old 3rd April 2011, 10:40 AM   #17
diyAudio Member
Join Date: Sep 2006
You could use the BD139 and BD140 for the VAS/current source stages with a small heatsink. BC546/BC556 won't hack it. Note that some simulators (and some models) don't model breakdown properly so won't tell you if you have a breakdown voltage problem.

  Reply With Quote
Old 3rd April 2011, 11:12 AM   #18
diyAudio Member
Join Date: May 2010
When you spell it out like that it makes sense!
I just measured my rails while turning the amp up fairly loud and they did droop by almost 3V even with 10,000uF of capacitance per rail (well 9400uF really, I prefer to use a number of smaller caps than individual big ones, its cheaper and gives less series resistance), and although I didn't actually measure the output, I probably realistically wasn't pushing it very hard.
A couple of questions though, when you say that the Vas/ current source can be drive to saturation, are we talking in terms of current saturation starving the bases of the driver transistors? If so, that could be prevented by using a high enough current in the class A VAS yeah?
I'll have to get myself a few MJE350's for current sources. I'm guessing the reason you don't recommend these as VAS transistors is low Hfe? Sounds like I'll also have to invest in a big 50V toroid if I want to make it to 100 real watts.
Also, do you know of any good sources for valve transformers? I've been thinking about building a valve amp for a while but it seems hard to find any reasonably priced locally.
  Reply With Quote
Old 3rd April 2011, 11:27 AM   #19
lineup is offline lineup  Sweden
diyAudio Member
lineup's Avatar
Join Date: Dec 2005
Location: the north
MJE350,MJE340 can take high voltage rails.
If you have +40/-40 Volt or lower then BD139/140 is better.
BJ139/140 are limited to 80 Volt operation.
Other VAS/drivers could be MJE15030,MJE15031
  Reply With Quote
Old 3rd April 2011, 11:32 AM   #20
diyAudio Member
Join Date: Nov 2006
Location: Adelaide, South Australia
Blog Entries: 45
The Voltage Amplifier Stage can go into saturation - by which I mean there is more current into the base than needed, and the Vce is at saturation (small fraction of a volt). Your schematic (from memory) included a transistor sensing the Vas emitter resistor voltage - this transistor is there in no small part to ensure that the VAS is not driven too hard. This is an important and often missed protection component.

--> Your current in the Vas is OK - increasing it for a 100W amplifier is not necessary.

MJE340 will "work" as a Vas transistor. The issue is more Cob and Ft. I have built amplifiers and used these - the effect is a bit of "sticking" on negative rail clipping (put one in and try it) and and an interesting effect on stability at high current near negative rail - which turns up as little bursts of oscillation at large negative excursions and high current.

BF469 / BF470 are older video transistors that are appropriate and quite likely available at your local shop.

The transformer you should be using for 100 Watts is likely to be about 36-0-36 VAC. Will rectify to close to 50-0-50.

I don't do valves. (Sorry tube heads!)
  Reply With Quote


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

Similar Threads
Thread Thread Starter Forum Replies Last Post
Randy Slone's "Fig 11.4" (Self's "Blameless"?): PCB layout tcpip Solid State 128 23rd September 2013 04:45 PM
I need comments about "100W power amp. elektor april 1982" jinkuchi Elektor 78 22nd February 2011 04:15 PM
Simplified LM3886 amp pcb "organic" style Groundloops Chip Amps 0 23rd May 2006 02:46 PM
Hybrid "SE Tube Amp Sound Style" Chip Amp idea ThorstenL Chip Amps 24 7th February 2003 06:41 AM

New To Site? Need Help?

All times are GMT. The time now is 03:47 PM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2