L20 V8 - Page 4 - 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 5th March 2012, 01:52 AM   #31
diyAudio Member
 
Join Date: Oct 2001
Location: Southwest
It's obvious these guys are not going to send or post a schematic, they are just making excuses and hedging their bets.
I learned my lesson from ordering an L7 without any documentation,just a bag of parts. Tried to get a schematic to no avail. Plus long shipping times
Never again will I order another Chinese amp kit, not worth the hassle and you can never trust the components.
If most of us stopped buying these kits without documentation they might get the message and do a better job.
Maybe the Moderators need to take a closer look at these guys.
__________________
Sam
  Reply With Quote
Old 5th March 2012, 02:40 AM   #32
AKSA is offline AKSA  Australia
diyAudio Member
 
Join Date: Sep 2001
Location: Melbourne, Australia
If they didn't sell so cheap, they would be able to spend the necessary time to do the documentation correctly and properly, in English.

There will always be a demand for the cheap product while this sort of shoddy outcome is tolerated. Unfortunately you do nearly always get what you paid for.

Hugh
__________________
Aspen Amplifiers P/L (Australia)
www.aksaonline.com
  Reply With Quote
Old 5th March 2012, 04:25 AM   #33
Bonsai is offline Bonsai  Taiwan
diyAudio Member
 
Bonsai's Avatar
 
Join Date: Jul 2003
Location: Taipei, Taiwan
I have some experience up at this power level, and if you obey the engineering guidelines, no class AB 250 watter is going to fit on a board the size indicated by the thread starter. If you bias up into class B, remove any form of protection, and are happy with sub-optimal slew rates and distortion, then you may get it. But, it aint hi-fi.

A 200W hi fi amp that can drive any reasonable hi speaker takes a lot of effort to design and build properly.

The best advice is to build something that's well documented and can be vouched for by other forum members (take a look at OStrippers designs), or buy a decent kit (AKSA). There's also the great Leach amp. Be prepapred to put quite some effort into the heatsinking as well. Class AB with .33 Ohm output degen resistors will need 78mA per pair - so for 4 pairs on 65V rails you are looking at an idle dissipation of 40W, and if you expect it to deliver 200W for any length of time, you should figure on heatsinks that will disspate 120-150W with typical speaker loads.

Now you know why decent 200-250W class AB commercial amps are so big and heavy and cost a lot . . .

;-)
__________________
bonsai
Amplifier Design and Construction for MUSIC! http://hifisonix.com/

Last edited by Bonsai; 5th March 2012 at 04:37 AM.
  Reply With Quote
Old 8th March 2012, 05:41 AM   #34
diyAudio Member
 
Join Date: Jan 2010
Location: Melbourne, Australia.
Default L20 V9

I finally tamed the L20 boards today. I fitted LPF's to each output and that put an end to the boards going into oscillation. I ran them both for about 5 minutes at 50WRMS but had to drop the power as the heat sinks got too hot. The dummy loads were sweating a bit too. I reconfigured the Cat 6 wiring I was using to connect from the amps to the dummy loads and this got rid of the rising frequency response and the overshoot on a 1kHz square way test signal. There is basically no DC offset on one amp and about 10mV on the other. There is some visible cross over distortion on a sine wave test signal and this may reduce if I can get the idle current up. At present each amp only draws about 3mA total at idle which seems to be far too low. All I have to do now is put together a speaker protection board and I can have a listen.
Attached Images
File Type: jpg DSCN2223.jpg (559.8 KB, 1092 views)
File Type: jpg DSCN2224.jpg (316.2 KB, 963 views)
File Type: jpg DSCN2225.jpg (595.8 KB, 940 views)
  Reply With Quote
Old 12th March 2012, 04:47 AM   #35
diyAudio Member
 
Join Date: Jan 2010
Location: Melbourne, Australia.
Default L20 V9

The Eagle has landed! I'm finally listening to the L20 V9 amp. It really sounds VERY good, especially since I removed the crummy RCA input cable and took my interconnect almost directly into the L20 board. There is some hum/buzz with my ear in the midrange driver but only when there is an input connected. Not so much a hum loop as direct radiation from the transformer/rectifiers. There is ony 3mA of idle current on each rail and some visible crossover distortion Any suggestions on how to increase the idle? I'd rather not experiment in case I blow it up (again).
Attached Images
File Type: jpg DSCN2233.jpg (459.7 KB, 925 views)
  Reply With Quote
Old 24th March 2012, 11:19 PM   #36
diyAudio Member
 
Join Date: Jan 2010
Location: Melbourne, Australia.
Default L20 V9

I obtained a further significant improvement in the sound by wiring an RCA socket directly to the input terminals and an even greater improvement when I wired the socket directly to the amplifier side of the input coupling capacitor (no DC from my source). Various previous L20 versions seem to have a small electrolytic as an input cap but my L20's came with a surface mount (stacked something or other) and it sounds terrible compared with the direct connection. Next trick is to wire up the 500VA toroids and see what +/-70V can do!
Attached Images
File Type: jpg DSCN2243.jpg (427.6 KB, 863 views)
  Reply With Quote
Old 25th March 2012, 11:40 AM   #37
diyAudio Member
 
Join Date: Apr 2010
Location: Coffs Harbour
As Bonsai has already posted, you are going to need significant bias current to get MOSFETs sounding good. The figure of 78 mA per pair may even be conservative, since the actual figure is quite rubbery. It's difficult to imagine how the sniff of just the few mA you mention, spread over four output pairs in class AB could be called bias. There is also the current flowing in the VAS (say 12 mA) and the input stage (say 3 mA) That's 15 mA without the output stage or any bias. I would think the measurement technique through, as something is wrong there. If the bias adjustment is actually fixed by the use of simple resistors to set it, there is obviously something quite wrong there too.

Surely, there is some adjustment for total bias current so it can be set to around the appropriate figure of about 360 mA. Yes, it's gonna get warm and with the higher rail voltages, very warm unless the heatsinks are quite generous. I would estimate sinks for 300W/4R to be better than 0.2 degrees C/Watt rated for each channel. In case you aren't aware, that's almost twice the size of the largest Conrad heatsink available here retail.

Don't try this amp with 70V rails on 4R loads at high power. It may limit itself but likely just wont survive on a stiff supply like the 500VA toroid.

Another concern is using a direct input connection with no DC isolation other than (hopefully) that which might be in the input source device. Frankly, I'm amazed you find the audio improved when the major crossover distortion problem at the amplifier output stage would obscure normal perception of audio quality. The capacitor supplied must have been sorely defective to be so obvious.

I imagine the heatsink is quite cold idling, but would you like to tell us what voltage (mV) you read across any of the 0R33 resistors?
__________________
Ian

Last edited by Ian Finch; 25th March 2012 at 11:44 AM.
  Reply With Quote
Old 25th March 2012, 01:11 PM   #38
diyAudio Member
 
Join Date: Jan 2010
Location: Melbourne, Australia.
Default L20 V9

OK, you guys are forcing me to do some head scratching, so here goes. The L20 V9 doesn't use mosfets in the output stages. It uses 2SD1047/2SB817 which are NPN/PNP planar silicons, 4 pairs. It uses a mosfet driver pair, IRF610/IRF9610 N-channel/P-channel. As supplied there is no adjustment for bias but I have put in a pot (shown on the attached schematic). As for my measurement of the current, I used a current clamp DC ammeter on both the +ve supply line and the -ve line. I got about 3mA on each. I managed to bias it up using the pot and thought that 100mA total might be a good number but naturally I blew up the amp (as per usual). Next time I took it to about 40mA total and as it heated up the temp compensation must have kicked in and the total current dropped to about 10mA. I then tweaked it a bit more and I think it is at about 20mA at present. As for the heatsink, 'IF' I was going to run them anywhere near full power I would use Conrad MF35-151.5 heatsinks which are good for 0.21C/Watt. I don't know if there is visible crossover distortion since I biased the output stages up a bit because I was so eager to listen that I did not put the CRO on it. But to my old ears it does not have any audible distortion. Believe me, the sound quality improved out of sight when I bypassed the input cap. When I find a good sounding cap I will put it in. The heatsink only gets slightly warm at any stage and as for the drop across the 0.22Ohm resistors, I haven't measured it yet. Sorry about the crappy sideways schematic. To be continued.
Attached Images
File Type: jpg IMG.jpg (135.4 KB, 708 views)
  Reply With Quote
Old 25th March 2012, 05:24 PM   #39
diyAudio Member
 
Join Date: Apr 2010
Location: Coffs Harbour
Ah, the light shines - it is not MOSFET so bias can be much lower. Apologies for the confusion but I'm not the first here to assume it was.

I think you will be led a merry chase expecting accuracy from a clamp meter at low DC current. Others may have a better appraisal but I would be sceptical, particularly in close packed circuits. For reliable readings, it's not difficult to measure the voltage across suitable resistors in the circuit, given that readings will be limited to the resistor+meter accuracies but you can expect ~3% across those Re resistors. A quick calc. in the head and it's done. You can then cross-check the reedings this way.

Let's see the real figures which, for ~25 mA/pair, should be 100 mA total, read as 0.22 * 0.025 = 0.0055 or 5.5 mV across each emitter resistor of the output pairs. For a little better chance of accuracy, read across both resistors in the pair, from emitter to emitter and expect ~11 mV with the same nominal 25 mA bias. Rest assured, if the sinks are getting warm with modest audio levels, the bias will be much more than 3 mA total. With typical setting, you might expect a total of 120 mA current drawn by each idle amplifier.

A standard 300 mm Conrad sink rated 0.37 Deg. C/W should be adequate, as the SC studio 350 uses it. The only problem being the close packing of transistors and local overheating. This is not good. BTW, I think a pot. is shown in the conventional position, in the base circuit of the bias generator transistor. There is some note there too but I can't make it out. Perhaps it specifies a fixed resistor in lieu?
__________________
Ian
  Reply With Quote
Old 25th March 2012, 11:31 PM   #40
diyAudio Member
 
Join Date: Jan 2010
Location: Melbourne, Australia.
Default L20 V9

I have scanned the schematic again and it looks a bit better. I also fathomed out how to rotate it. The clamp meter I am using is of professional quality and I have no reason to doubt it as it has agreed with other measuring methods in the past. Still, I will now measure the voltage across the 0.22 Ohm resistors as I have always done with other amps. I was wary of poking the meter leads into this PCB in light of it's previous liking for 'taking off'. The pot shown in the circuit is one I put in. I used 500 Ohms at first but changed to 5K when I didn't get much response from 500. Also, now I look at the circuit in more detail I see that there is some protection against HF signals (or have I got that wrong too?) in the form of a 10 Ohm resistor and 0.1uf cap to ground on the output. This would represent (?) a 170 ohm load at 10kHz and 26 ohm load at 100K. I suppose it could be construed as being there to stop stuff getting back in from the speaker wires as much as loading the amp. I will have to get Mr. Selfs' book out again and have a look. If the amp went into oscillation I suppose it would make mincemeat of the 10 Ohm resistor and maybe it has. Regards.
Attached Images
File Type: jpg IMG_NEW.jpg (609.7 KB, 721 views)
  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
L20 AMP。use only two NJW0302G ljm_ljm Solid State 78 13th July 2014 07:22 PM
ebay amp L20 power amplifier meko Solid State 54 15th April 2014 09:08 PM
Has any one tried this l20 - 350 watt amp srinath Chip Amps 1 12th January 2011 07:10 PM
V8 Fonken design ? Bigun Full Range 30 28th January 2009 09:58 PM
Fast tractor - V8 peranders The Lounge 1 15th October 2008 03:47 AM


New To Site? Need Help?

All times are GMT. The time now is 09:01 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