1000 Watt Sub Amp: Design / Build - Page 2 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

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
Reply
 
Thread Tools Search this Thread
Old 24th March 2008, 05:47 AM   #11
GK is offline GK  Australia
Account disabled at member's request
 
Join Date: Jan 2006
Also, looking at your circuit a little closer, I think that more than three diodes would be required for class AB operation in real life and much larger emitter resistors will be needed for the output pairs. The spice models for your driver and output transistors are likely returning an unrealistically low Vbe. On Semi's models for these transistors are really bad and shouldn't be relied upon.

0.1 ohm at 85mA bias per device with such high voltage rails may be thermally unstable. 0.47 to 1.0 ohm ballast resistors would be better.
Since this is only a subwoofer amp, the increased crossover distortion caused by gm doubling with large emitter resistors isn't something to worry about as much, as there is HEAPs of loop gain and global negative feedback throughout the frequency band of interest.
  Reply With Quote
Old 24th March 2008, 10:26 AM   #12
GK is offline GK  Australia
Account disabled at member's request
 
Join Date: Jan 2006
Below I've attached the junction voltage drop curves for your biasing diodes, the output transistors and the driver transistors.

With a VAS quiescent current of ~20mA each IN4001 diode will give you a little under 650mV bias voltage. You are looking at about 600mV for each driver and output power transistor, so a biasing voltage of over 2.4V will be required to develop a standing voltage in the power transistors for class AB operation. Three diodes will give less than 2V.
Attached Images
File Type: gif vbe.gif (42.9 KB, 3921 views)
  Reply With Quote
Old 24th March 2008, 10:18 PM   #13
wg_ski is offline wg_ski  United States
diyAudio Member
 
Join Date: Oct 2007
Default Re: Re: Re: 1000 Watt Sub Amp: Design / Build

Quote:
Originally posted by MJL21193


I did a fair bit of research on heatsink requirements for this amp. Using this formula: Pd = (Vpp^2 )/(R(Load)*2 * PI^2 to determine total dissipation at full power I get 490 watts at 2 ohm load. Lotsa heat!

Here's the heatsink, with the 18 output devices laid out to see how they fit.
Plenty of heatsink when used with a small quiet fan. It's about the same size I used in a CS800 clone, with two clamshelled together. It runs 1.4 ohms per side with 7xMJ21193/4 on +/-80v and never has to switch to full speed on the fans.
  Reply With Quote
Old 24th March 2008, 11:16 PM   #14
Dr_EM is offline Dr_EM  United Kingdom
diyAudio Member
 
Dr_EM's Avatar
 
Join Date: Oct 2006
Location: Swindon
Looks like an interesting project! I like the idea of so many output devices, seems much more solid . The heatsink looks to be designed mainly for convection cooling since the fins are large and quite well spaced but a fan should still increase its effectiveness a fair amount. How will you mount the semis? I'm going to use M3 allen head bolts on my new amp, but clamp mounting is my favourite where possible.

You'll need some really thick cabling and solid grounding, just how much current can you put through a normal copper clad board? I usually use them for my power rails (storage caps) and ground for amps (cuts to form 3 sections). A thicker piece of copper might be better at this power level.
  Reply With Quote
Old 25th March 2008, 12:43 AM   #15
diyAudio Member
 
Join Date: Nov 2004
Location: Florida
Thick copper would work, that's for sure. For 1000W @ 2 ohm he's probably needing the lowest resistance paths possible to handle the current!

When I built my 350W sub amp, I used thick wires and PTP (Point to Point) soldering on the power devices and drivers because of the high currents. I used the PCB only for the lower current VAS and differential stages, then used this circuit to "drive" the output stage.
  Reply With Quote
Old 25th March 2008, 12:46 AM   #16
diyAudio Member
 
Join Date: Nov 2004
Location: Florida
Quote:
Originally posted by MJL21193



Hi EWorkshop1708,

The problem with raising the VAS current in the present configuration, is that idle current will also increase. The VAS is running with ~12mA right now. I'd like to increase it - up to 20mA or more (the device can handle it), but that means either removing one diode and let the amp run B or ditch the diodes and use a Vbe multiplier.
You can put a resistor in parallel across the diodes to drop the idle current, and have a higher VAS current.
  Reply With Quote
Old 25th March 2008, 01:35 AM   #17
Account disabled at member's request
 
MJL21193's Avatar
 
Join Date: Mar 2007
Quote:
Originally posted by G.Kleinschmidt

An opamp based low pass filter preamp with selectable turnover frequencies and roll off slope for this amplifier would also be something to look into.

PS, with a 1kW subwoofer amplifier Iím glad youíre not my neighbour.
Hi Glen,
Thanks for your interest here. This amp will get signal from the Linkwitz transform which is fed from the subwoofer output from my X-Fi extreme audio sound card. This sub is almost exclusively for home theater, with my HTPC being the source.

Come now! I think we would get along well as neighbours.
Besides, it's only 1kw at 2 ohms, a meager 500 watts at 4. Didn't you build a 1kw class A Amp?

Quote:
Originally posted by G.Kleinschmidt
Also, looking at your circuit a little closer, I think that more than three diodes would be required for class AB operation in real life and much larger emitter resistors will be needed for the output pairs.
I'm scrapping the diodes in favour of a Vbe multiplier - more room for adjustment. Less hassle mounting a TO-220 to the heatsink than three diodes.
The .1 ohm emitter resistors is leftover from when this amp design was class B, I just didn't think to change these.
Fact is I don't have enough 5 watt emitter resistors of any value on hand to complete this, so I'll be picking up some more. Could go .33 ohm, like is in my P68.

Quote:
Originally posted by G.Kleinschmidt
Below I've attached the junction voltage drop curves for your biasing diodes, the output transistors and the driver transistors.

With a VAS quiescent current of ~20mA each IN4001 diode will give you a little under 650mV bias voltage. You are looking at about 600mV for each driver and output power transistor, so a biasing voltage of over 2.4V will be required to develop a standing voltage in the power transistors for class AB operation. Three diodes will give less than 2V.

Like I said above, I'll be going with a Vbe bias, but thanks for the info.
Multisim doesn't realistically simulate Vbe (due to inaccurate models, like you say?), as I've found before. The values for the resistors in the bias servo in the simulation doesn't give the correct results in the real circuit. To simulate is one thing, but to get the actual values I need to calculate. So, the values shown on the schematic now are for simulation only and will be changed to the correct ones in the actual amp.

Here's an updated schematic. Includes the Vbe multi and the emitter resistor value changes. Also I've increased the VAS current to ~27mA.
Attached Images
File Type: gif 1000 watter.gif (32.1 KB, 4291 views)
  Reply With Quote
Old 25th March 2008, 02:00 AM   #18
Account disabled at member's request
 
MJL21193's Avatar
 
Join Date: Mar 2007
Default Re: Re: Re: Re: 1000 Watt Sub Amp: Design / Build

Quote:
Originally posted by wg_ski


Plenty of heatsink when used with a small quiet fan. It's about the same size I used in a CS800 clone, with two clamshelled together. It runs 1.4 ohms per side with 7xMJ21193/4 on +/-80v and never has to switch to full speed on the fans.
Hi wg_ski,
Yes I want to incorporate a temperature controlled fan that will stay off until the heatsink temp gets to ~70 degrees C. This may be insurance more than a requirement at the higher impedance.


Quote:
Originally posted by Dr.EM
Looks like an interesting project!

How will you mount the semis?

You'll need some really thick cabling and solid grounding, just how much current can you put through a normal copper clad board? I usually use them for my power rails (storage caps) and ground for amps (cuts to form 3 sections). A thicker piece of copper might be better at this power level.

I have been thinking about mounting the outputs on the heatsink and wiring them direct, without a board. Use 12 gauge solid copper wire for the rails and output. this would simplify things and I could put the rest of the circuit on a smaller board, with the drivers and the Vbe multiplier transistor on the outer edge to mount direct to the heatsink.

Another possibility is to cut the heatsink down the middle and direct couple the outputs to the sinks, using those as the voltage rails. This would mean having the sinks isolated from the chassis and moving them inside the case to avoid body contact with the potentially lethal voltage
That would mean I'd definitely need fans for cooling.

Possibilities!
  Reply With Quote
Old 25th March 2008, 02:17 AM   #19
Account disabled at member's request
 
MJL21193's Avatar
 
Join Date: Mar 2007
Quote:
Originally posted by EWorkshop1708
Thick copper would work, that's for sure. For 1000W @ 2 ohm he's probably needing the lowest resistance paths possible to handle the current!

When I built my 350W sub amp, I used thick wires and PTP (Point to Point) soldering on the power devices and drivers because of the high currents. I used the PCB only for the lower current VAS and differential stages, then used this circuit to "drive" the output stage.
Yes, this is what I had in mind. Pretty much a waste of a circuit board to mount that many devices.
What do you think of my idea to use the heatsinks as the rails? The more I consider it, the more I like it.

Quote:
Originally posted by EWorkshop1708


You can put a resistor in parallel across the diodes to drop the idle current, and have a higher VAS current.
Thanks! I didn't know this. One of the hazards of being a rookie.


  Reply With Quote
Old 25th March 2008, 03:11 AM   #20
diyAudio Member
 
Join Date: Nov 2004
Location: Florida
Quote:
Originally posted by MJL21193


Yes, this is what I had in mind. Pretty much a waste of a circuit board to mount that many devices.
What do you think of my idea to use the heatsinks as the rails? The more I consider it, the more I like it.


It's fine, just don't touch the heatsinks during use Since you will have no thermal pad, there will be good thermal contact. It's a hassle to cut heatsinks however. Since it's a huge amp, your EF topology is fine.

What I did on my amp was use CFP and no insulators, and the NPNs (+ PNP driver & silpad) on one heatsink, and the PNPs on the other sink. Since both sinks are near output potential, I get no shock, unless I touch ground and the heatsinks while the amp is playing loud.
  Reply With Quote

Reply


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
quasar amplifier by dr bora 1000 watt andrewlebon Solid State 64 27th October 2013 10:35 AM
1000 watt amplifiers boxcustom Car Audio 32 31st May 2011 04:49 PM
Help With A 500 Watt subwoofer and a 1000 amp Bgpoet Solid State 2 30th December 2008 07:41 AM
1000 watt Amp Modules Zero Cool Class D 24 29th September 2008 08:32 PM
1000 Watt with usb input ! cph2000 Class D 1 4th October 2007 06:04 AM


New To Site? Need Help?

All times are GMT. The time now is 04:07 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2