amplifier troubleshooting question

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

I am currently building a clone of the M250 pictured here: http://www.astro.uu.se/~marcus/private/m250.html

but have stumbled into some difficulties. At first I build the front-end (everything but the 8 transistors in parallel and everything worked perfectly. I applied a 1khz signal to the input and got a beautiful replica of the 1khz signal on the output. Once that was working I decided to connect the front-end to the current amplifier but when I applied power the current ballast resistors started smoking. I am using 5W .33 OHM resistors like the schematic shows but no luck. While I’m doing this test I have no signal feeding the input nor do I have a speaker or dummy load connected across the output. Any ideas would be greatly appreciated!

Thanks,
Randy
:scratch:
 
When first testing, USE A LIGHTBULB in series with the AC line. No kidding! This is a life saver. You will have to make up your own lightbulb socket, but make sure that it is in SERIES of the amp. This is a very old technicians' trick. Think about it: A 100-250W 120 volt bulb will have a very low COLD resistance. If the needed idle current is reasonably low, the lightbulb will stay relatively off and still have a low resistance. However, if the amp requires lots of idle current ( a dangerous condition) then the bulb will heat up and become high resistance, limiting the absolute current through the whole amp. It makes the amp last much longer, before breaking and usually you can even determine what the problem is, while still running the amp (sort of). I hope that this helps.
 
Hey Randy,

I wish I could see the damned schematic. It's too big for my poor video card I guess. 32MB video card should be enough shouldn't it? Oh well..... Check for DC on the output. What polarity is it. This sometimes gives clues where to look. The lightbulb trick is a good one or a Variac will do also. If no DC on the output, check output for oscillation. If oscillation, put a 100pF cap from base to collector of your drivers.... This will get you there quicker. If you set the bias without the output transistors in circuit, this will cause big biasing problems. Turn the bias all the way DOWN, connect the output stage, and slowly power the unit. Sorry I can't be more helpful. I can't see the diagram.....

Chris
 
Hi,

Suggest you check your wiring VERY carefully. As there is no connection between R22, T8 or T9 and the output line when the output transistors are not connected, you should not have seen a signal on the output line during your initial test.

Suggest you remove the output transistors and connect two 100ohm 1W resitors, one from T8 emitter to ouput line and one from T9 emitter to output line. You should see a maximum of 0.6V DC across each of these resistors, and 0V DC on the output line. The pot P1 should adjust the voltage across the extra resistors. If all ok here, amplifier should amplify input signal with no load.

let us know results of this check

Cheers
 
I would check the setting of P1, which generates the bias voltage for the output stage and thus controls the its idle current. P1 should be set to its maximum value. If it is set to its minimum value, the idle current will be excesive and could blow up the output stage. To adjust P! you will need to put an ameter between one of the power supply rails and the amp. Decrease the resistance of the P1 until the desired idle current is achieved. Let it run for a while to stabalize thermally and then reset it if the current has drifted. It may take several itterations to get it stabalized at the value you want.
 
Greg B. said:
I would check the setting of P1, which generates the bias voltage for the output stage and thus controls the its idle current. P1 should be set to its maximum value. If it is set to its minimum value, the idle current will be excesive and could blow up the output stage. To adjust P! you will need to put an ameter between one of the power supply rails and the amp. Decrease the resistance of the P1 until the desired idle current is achieved. Let it run for a while to stabalize thermally and then reset it if the current has drifted. It may take several itterations to get it stabalized at the value you want.


indeed I guess this is the problem! :nod:

there's also a thread about this amp: link

best luck!

HB.

EDIT: I remember this amp has a short-circuit protection, why isn't it working in this case?? :scratch:
 
Hey, Who Said I Am An OLD Technician...

The lightbulb trick is a good one or a Variac will do also.
Use both with a 60W lamp and power one channel at a time for initial no load testing.
Up the lamp wattage when connecting a load - if you have cross-conduction or oscillation problems the lamp will limit the fault currents and save your expensive output transistors.

Eric.
 
Dear old technician, I am one as well, from so many decades ago. ;) I learned this trick from a REAL technician, and sometimes I forgot about it, over the years. Well, once, in 1978, I had a new amp design that I had to make it work for a CES. It ALWAYS blew up. In frustration, I hired the tech who had trained me and he brought his own light bulb assembly. Guess what? My personal tech, FORGOT to put the transistor insulators between the driver transistors and the heatsink. We found the problem and it went to the show.
 
Damned techs....

Yup, I know the feeling. At least he got the polarity right. Sometimes I get the polarity wrong especially when there is no silkscreen and the mechanical drawing is ****-poor. (Galian-Kruger) is terrible about that as well as OLD Peavey stuff! Oh well, Sometimes it is my job to let the smoke out of parts that have too much smoke pressure. My boss once told me "Well, there ain't any ****** left in it anymore...." I scratched my head and he said "Yeah! you blew the ****** out of it!" Funny-guy, funny-guy........ He always says that when people blow up their speakers and are in for re-cone.

Chris
 
Oldie But A Goodie....

Hi John,
Yeah I fix a lot of big Pro PA amplifiers, and I do not dare to run them up after a rebuild without the Lamp/Variac current limiting - if all those output transistors go splatt I get to buy a new set out of my own pocket and that blows any profits on the job.
I also have a very large Variac (3kVA) and this I use to run up things like Krell and other high current Class A amps.
I have never left the insulators out, but a fine solder bridge left can wreak expensive havoc, and I am not interested in the anxiety factor when running up a new or repaired amplifier.
Also these big amplifiers are needed as soon as they are repaired, and a few days delay getting more parts can be a problem too.
Glad to know that I am not the only 'old tech' who makes the rare mistake. ;)

Eric.
 
Faulty Technician ?.

Diode said:
Damned techs....

Yup, I know the feeling. At least he got the polarity right. Sometimes I get the polarity wrong especially when there is no silkscreen and the mechanical drawing is ****-poor. (Galian-Kruger) is terrible about that as well as OLD Peavey stuff! Oh well, Sometimes it is my job to let the smoke out of parts that have too much smoke pressure. My boss once told me "Well, there ain't any ****** left in it anymore...." I scratched my head and he said "Yeah! you blew the ****** out of it!" Funny-guy, funny-guy........ He always says that when people blow up their speakers and are in for re-cone.

Chris

Hi Chris,
I use a black marker pen to denote transistor types, connectors etc before I remove them.
I find this is quick, easy and saves time, money, trouble and grief.
And also going to your boss and saying " Ummm, I am an idiot and I blew the amp up AGAIN, and I need more new parts".
Where I come from there is NO excuse for making mistakes like these, and in my younger days a mistake like this meant getting the coffees and lunches for the whole workshop for the next week, as well as the drubbing so richly deserved.
A workshop chant at the time was the six P's - Prior Preparation Prevents A P*ss Poor Performance.

You have been told ;) ,
Eric.
 
check t18

you could have reverse connected it, the emitters of t8 and t9 are connected to each other thru r22 and c6 and as per you schematic is not connected to the output line so i was wondering how you were able to get a 1khz signal out at all....or maybe you just connected the base and emmiters of the outputs but not the collectors, so in this case it iwas possible you got a signal out...is this what happened?
 
Hey Eric,

Yeah, I do make the occasional mistake, especially when being hurried. I'm not a good hurrier and I hate pressure. I also work for a good guy who knows the price of doing business. He also knows that its just life and it will happen time to time, even from himself. Hell, If I had to do the things you mentioned, I would have left the business long ago and did something else. The pressure of rebuilding a 2000W amp knowing you just dropped a couple hundred dollars into the final stage alone, mixed with the consequences of being out of pocket the money for lunches for a week, I wouldn't tollerate and I never would have learned a lot of valuable lessons and gotten where I am today. You are right though.... A couple times of that and repairs would be a lot more thorough. I'm not being paid hourly and time is a luxury I don't have else bills don't get paid. When I DID get paid hourly, I was always behind schedule by a month, well, at least that's what was said in meetings...... BOSSES, you can't please em..... I don't believe in perfection but some people are just plain better or luckier than others. I mark things where and when I can but sometimes it just ain't your day........ Interesting bit of culture there..... Thanks for sharing!

Chris
 
Hi Chris
You got me slightly wrong there.
In the case of self employment (current situation) if I make a mistake and blow something, I get to pay for the new parts and that erodes any profit I was meant to make on the job - all the more reason to be meticulous.
In the case of previous employment, if I blew something I did not have to pay except for having to admit my mistake.
The workshop punishment was to be the lunchboy, which meant going to the local lunchbar and collecting everybody's orders, but not paying for them.
Aussie culture is such that if a mistake is made in a group situation, the group gets to know about it, and the offender is reminded of their stupidity - nothing personal.

Eric.
 
just got it working!

Hi All,

I'm not using a transformer just several car batteries in series. Last night I figured out what I did wrong. I accidently shorted the base to the collector on one of the output transistors and that resulted in my problems. I couldn't even see the damn short but my meter registered it. As soon as I get everything squared away I'll post some pictures.

To joan2: Yep, I connected the feedback loop to the output of the driver transistors so it would work without the output transistors.

Thanks again!
Randy
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.