Go Back   Home > Forums > >

Solid State Talk all about solid state amplification.

ETI 477 article
ETI 477 article
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
Old 13th November 2020, 10:32 PM   #111
owdeo is offline owdeo  Australia
diyAudio Member
 
Join Date: Jun 2003
Location: Sydney
I've repaired a few 5000 amps over the years and all of them had failed/failing BC550 input devices. They are probably running too close to their maximum voltage ratings in this design and hence do not last as long as they might. I replaced them all with (matched) BC546 which has a higher rating and found this to work well. The BC550 is a lower noise device, but in practice I don't think there was any noticeable difference sonically. I also found the power supply and grounding of two amps I repaired was not wired according to the original article and would not have been optimal -depends on the competence of the original constructor of course.

These are great amps in my view - a real classic. When constructed and working properly and brought up to date with modern components that is...also they need a bit more bias in the output MOSFETs than the original spec to sound their best. They have a certain character that really suits 80s pop, but also very clean and unstressed at high output levels. They can play a full symphony orchestra at maximum climax without sounding hard or stressed. Love the front panel heatsink too, mind you it isn't really up to the job... Interestingly the Electronics Australia "Pro Series 1" (and later PS3) that supposedly replaced and improved on the 5000 sounded awful and was a far more basic design with higher distortion levels.
  Reply With Quote
Old 13th November 2020, 10:38 PM   #112
owdeo is offline owdeo  Australia
diyAudio Member
 
Join Date: Jun 2003
Location: Sydney
PS I think from memory you need two 18V zeners in series to balance out the power dissipation in the non-driving side of the diff amp VAS transistors. I never got around to trying this mod - the amp works well without it. A common base transistor with the base grounded would probably be a better way of doing it, if potentially less tidy to add to the PCB.
  Reply With Quote
Old 14th November 2020, 03:40 AM   #113
Ian Finch is online now Ian Finch  Australia
diyAudio Member
 
Join Date: Apr 2010
Location: Coffs Harbour, NSW
ETI 477 article
In its original form, at least one seller of ETI5000 kits supplied woefully inadequate thin aluminium strips for the VAS stage heatsinks, where larger "L" section extrusions were specified. From memory and the prompting of a tiny schematic, there was an MJE340/350 pair and 4 X BF469/470 video driver transistors in each channel VAS. It was an elaborate design and without enough cooling, the transistors that normally ran hot could eventually fail.

My own amplifier was a fancy kit version kit by Jaycar, with beryllium oxide TO3 insulators and no, I wasn't poisoned by prolonged exposure to any dust because there wasn't enough to see or remove with adhesive tape to examine under a microscope. (I was a lab. staffer and tech. at the time) It eventually worked fine but doesn't, last I checked, sound as good as I would expect of specialised semis like Hitachi's lateral mosfets in an advanced circuit like this one.
  Reply With Quote
Old 14th November 2020, 08:16 AM   #114
Leinster Lad is online now Leinster Lad  Australia
diyAudio Member
 
Join Date: Jun 2013
Location: Perth
Still not much luck.


Tried a bunch of BC550's. hfe and Vbe matched within a mV or two.

Minimal difference across many different transistors.
I must have tried 10 different "matched" pairs. Nada.


This issue cannot be caused by the input diff pair.



Tried the trim pot as suggested.
Minimal (maybe 5mV) difference until at full stroke, which effectively shorts out the 270ohm emitter resistor of Q1, which I had discovered reduced the offset voltage earlier.


This indicates that whatever the "unbalance" is, it is pretty substantial, and possibly in the VAS ?



Just picked up some zeners so going to fit and test.


Could it be the mosfets ? one side conducting a touch more than the other ?


Bit strange that two modules would have identical failures.
  Reply With Quote
Old 14th November 2020, 08:22 AM   #115
Leinster Lad is online now Leinster Lad  Australia
diyAudio Member
 
Join Date: Jun 2013
Location: Perth
Quote:
Originally Posted by owdeo View Post
they need a bit more bias in the output MOSFETs than the original spec to sound their best.

Hi Owdeo, what idle current do you suggest ?
I have heard that 50mA thru each pair is minimum and some have suggested going as high as 100mA per pair ( 200mA per 477 module )




Quote:
Originally Posted by owdeo View Post
A common base transistor with the base grounded

Could you detail this please. It might be worth a shot ?
  Reply With Quote
Old 14th November 2020, 08:28 AM   #116
Leinster Lad is online now Leinster Lad  Australia
diyAudio Member
 
Join Date: Jun 2013
Location: Perth
" Look at the asc of my version of eti477. "


?? cannot open .asc files. Would you have a PDF or JPG of you circuit ??




"increase gate resistors and use there Dale/Vishay RN65s."



I have heard to increasing to 220ohm ???

What does this achieve ??




"Add 2k trimpot between LTP emitter resistors. Enough space on the pcb. That will allow you bring offset to below 1mV. "


This did not work - until the Q1 emitter resistor was effectively shorted by the pot. Offset voltage then came down to only 18mV




You may separate PS to output mosfets and VAS with diodes or 10 ohm/2-5W resistors.


Easy to do, What are the advantages of doing this ??
  Reply With Quote
Old 14th November 2020, 08:31 AM   #117
janusz is offline janusz  Australia
diyAudio Member
 
Join Date: Oct 2004
Location: Perth
Hitachis are very robust. I'd check VAS first as their high temperature must have damaged pcb - at least to some extent so possibly it conducts a bit - or/as well somewhat damaged VAS transistors due to overheating.

It's worth replacing BC550s with BC546s as suggested above or some other ones such as 2sc2240s I have used or eg. ksc1815.

cheers,
  Reply With Quote
Old 14th November 2020, 10:16 AM   #118
Leinster Lad is online now Leinster Lad  Australia
diyAudio Member
 
Join Date: Jun 2013
Location: Perth
Hmmm. PCB board only has minimal discolouration, so I would say that there is no damage to the board.
This amp had very decent heatsinks on the VAS which is why the PCB is still good.



BF's may be mismatched / degraded, so plan to replace with ksa1142/ksc2682. Of which I have a few spares.



The old 550's test good and fitting new matched 550's made no difference to the issue at hand.
  Reply With Quote
Old 14th November 2020, 10:21 AM   #119
Ian Finch is online now Ian Finch  Australia
diyAudio Member
 
Join Date: Apr 2010
Location: Coffs Harbour, NSW
ETI 477 article
The VAS is where the little problems become big ones. For starters, make sure you can't accidently slip and/or short between the transistor pins and any rail with (blunt)DMM probes. Use sharp points, and hook probes or at least small clip leads for common return points as appropriate. Then check all VAS base-emitter voltages on the TO126 transistors when there is no signal input or speaker connected.
You can also check base-collector diode voltages but it isn't often as telling. You should read a diode voltage of around 0.68V - say, 0.7V to allow for measuring differences, across either junction, either polarity. If this proves to be consistent and OK in both channels, check the overall VAS stage currents by measuring the voltage drop across their power supply resistors of each stage. The VAS currents should be fairly consistent in both channels.

As far as output stage bias current is concerned, mosfets of all kinds deliver their best figures at maximum current, as in full class A operation and lots of heat. Class AB operation is thus a sliding compromise of efficiency, heat and performance where there is no defined optimum bias level, just a need for as much as is reasonably possible. Fortunately, things do begin to sound good when the amplifier is biased up to around 100mA per output pair, which is quite warm when the rails are above +/- 40V. This is about twice the bias level ETI 5000 was designed for and the front panel heatsink will be roasting in summer. Perhaps you can leave that type of project 'til later, when this problem is located.

Last edited by Ian Finch; 14th November 2020 at 10:24 AM.
  Reply With Quote
Old 14th November 2020, 11:50 AM   #120
Leinster Lad is online now Leinster Lad  Australia
diyAudio Member
 
Join Date: Jun 2013
Location: Perth
Thanks Ian.
I was playing with the mosfet bias just to see if it affected the output offset voltage.. it did not.
Will definitely play with them though once I get this issue solved.


I did already check voltages across power supply resistors on each stage and they were basically exactly as the schematic describes, which puzzled me as I was sure that I was going to find some clues there.


I was wondering if there was any kind of servo action happening via the NFB because the trimmer pot across the diff input emitters basically did nothing until the pot was at end of travel and shorting out one of the resistors.


Could a leaky millers cap ( C14, collector base of Q6 ) or leaky C4 ( between bases of Q3 & Q4 ) do this ?
  Reply With Quote

Reply


ETI 477 articleHide this!Advertise here!
Thread Tools

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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
EDN article Claude Abraham Power Supplies 1 14th October 2006 05:57 PM


New To Site? Need Help?

All times are GMT. The time now is 12:48 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Copyright ©1999-2021 diyAudio
Wiki