infinity overture 2 repair

[Jason Kooner]

Hello,

I have a pair of infinity overture 2 speakers. They have an internal amp to power the woofer section. However, I don't here anything from the woofer. There is a midrange and tweeter that are not powered and they work fine.

I unplugged one of the speakers and opened it up. There is some black dust on the pcb near the power fets. There are 4 fets and they are all shorted. The fets are Fairchild IRFS250B. I couldn't find these (too old) so I replaced them with ST Micro STW40NF20 fets. Also the power fets are grouped in 2 pairs with one pair having an NPN transistor between them, why doesn't the other power mosfet pair have this NPN transistor? The NPN is a Fairchild 2012, it measures ok with a meter but I went ahead and replaced it anyway.

I plugged in the speaker and tested it. The woofer is alive now but sounds terrible, it's all distorted. The midrange and tweeter are fine.

I plan to bring a scope and thermal camera home from work for the weekend to take some more measurements.

Questions:

1. Was that wrong to sub in the ST Micro fets? I did a datasheet comparsion and didn't see anything special about the Fairchild fet.

2. Does anyone have a schematic for this pcb?

3. What other parts should I look at?

I think I was too quick to clean the black dust and should off spent some more time trying to figure out the source!

Cheers!

-jason

[chuck]

Check for open resistors and open drive transistors.

[Mark Zacharias]

"chuck" <ch...@deja.net> wrote in message
news:ag1bj9ld72u9vvplr...@4ax.com...

UGH. Had to work on one of these once. Reminded me of a car amp.
Don't forget to use an isolation transformer.

The manual can be downloaded from Elektrotanya or you can email me at:

mzach...@labolgcbs.net

and reverse "labolgcbs" to read "sbcglobal".

They do get a number of open resistors. I my case, a predriver looked good
but wasn't. Caused additional heartache. Got it working, but it wasn't worth
the trouble for me. If business hadn't been slow I would have just turned it
away.

Mark Z.

[Jason Kooner]

I have the manual now ...thanks. Also I did find some 0.22ohm half watt resistors that were open. There is also one BF469 NPN transistor that is bad but I can't find it online. I tried searching for recommended alternate but no one carries those either...too old.

Can I just lookup the Hfe, voltage and ft on the datasheet and search on mouser od digikey for the same specs? Or does anyone know of a BF496 sub that I can get in 2014?

thanks,

-jason

[chuck]

On Mon, 31 Mar 2014 10:19:55 -0700 (PDT), Jason Kooner

NTE108. Mouser Electronics or MCM may have it in stock.

[Jason Kooner]

I replaced the bad components but it failed again. I think I missed a few 1K fusible resistors that are open. They connect to the gates of the power fets so I think the gate was floating on 2 fets when I fired it up.

I'll replace all the bad parts again but this time I'll bring it up slowly at my work since I have better equipment there.

I do have 2 questions:

1. On page 32/35 of the schematic there is a 2K pot in the near the middle of the page. I assuming this is used to adjust the bias of the power stage to some level. I have a function generator to help with this, where should I place my scope probes when adjusting this pot and what am I looking for?

http://elektrotanya.com/infinity_overture_2.pdf/download.html


2. On the same page there is a 1K 2W (R68) resistor on one of the relay contacts, what is this for?

thanks,

-jason

[chuck]

On Sun, 6 Apr 2014 20:59:26 -0700 (PDT), Jason Kooner

I would replace all transistors Q 8 to Q 16. Make sure none of the
fused resistors are open. Check all zeners for leakage and shorts.
Also D11. Put a 60 to 100 watt light bulb in series with the hot side
of the AC input and, once the light bulb dims and the relay closes,
plug the amp directly into the wall and measure bias across one of the
.22 ohm resistors on the output fets. (Infinity specifies R77) Adjust
for 5.9 mv after running the amp for 15 minutes. The pot is a bias
pot. The R68 resistor appears to drain off the offset voltage when
the amp powers down. This design is at least 30 years old so there
shouldn't be any issues that Mark Z or I haven't seen before.

[Jason Kooner]

- I replaced Q8-Q16.

- I subbed 2N5655G for the BF469 and MJE350STU for the BF470

- I powered it up using a DC bench supply with the current limit set to about 1A

- The power fets were getting very hot.

- I checked more parts, R74 shold be 36ohms and measured 36K so I changed it.

- I brought the DC input up to 160V and the power fets are not getting hot.

- The input power is about 4W


Questions:

1. The Vgs of the high side fets (Q13,Q14) is about 2.8V and the low side fets (Q15,Q16) are 2.0V, shouldn't they all be the same?

2. Before I plug this into the mains I want to make sure I know how to bias this right. Chuck, you mention setting R77 to 5.9mV, what about measuring the output voltage (Q16 Drain) and set it to 80V (half the DC input)?

3. Looking at the schematic, I can see the 160VDC bus is not regulated. Won't the bias conditions change if the 160VDC changes? Do newer designs use a regulated supply?

4. This design used 4 NFETSs instead of 2 NFETS for the low side and 2 PFETS for the high side. Is this for cost? Are newer designs done this way?


thanks,

-jason

[chuck]

On Sun, 20 Apr 2014 12:03:22 -0700 (PDT), Jason Kooner

I wouldn't worry about issue 1 just yet. Power the unit up using the
light bulb technique that I mentioned in a previous post.

Issue 2. You need to set the bias the way I said but you don't set
the bias before you ascertain that the amplifier is working properly.
When the light bulb is still in circuit, inject a low frequency sine
wave into the input and watch the output for symmetrical clipping.

Issue 3. Amplifiers have been made with unregulated supplies for
generations. After the amplifier warms up, set bias with the AC input
around 120 V AC and you should be good to go.

Issue 4. I have no idea. The schematic I am looking at has a total
of 4 fets 2 top 2 bottom. If I was to guess, Infinity probably used a
lower current fet for the bottom so twice as many were necessary
although I have never seen that before and I have been repairing amps
since 1971. Chuck

[Jason Kooner]

Chuck, thanks for you help I think it working better now. I did read the service manual and I see the part about the 5.9mV.

I inserted a light bulb in series with the hot side and adjusted the pot until the light bulb shut off. So, the light bulb is needed to adjust the pot to prevent shoot-through current when the amp is plugged directly into the mains?
I then plugged the amp directly into the mains and adjusted the pot for 5.9mV on R77.

Next I put the bulb back in and injected a 30Hz signal. I increased the signal until it clipped but it's not symmetrical, it starts to clip at the bottom first:

https://www.flickr.com/photos/124747932@N05/14259458203/


do you think there is still something wrong? I did connect it the speaker and it sounded ok to me.


I also took a thermal scan of the pcb without any load on the output and noticed one hot component. The 5.1K resistor (R23) used in the 24V supply reaches 157C. Power dissipation was measured at 3.2W

https://www.flickr.com/photos/124747932@N05/14052632878/


The pcb is discolored around R23 so I'm thinking of replacing it with a TO-247 package so it can mount to the frame just like the power fets.

I started to troubleshoot the second amplifier and noticed that the pcb is slightly different from the first amplifier. There is note about this in the service manual. The second board has a MPSU06 transistor for Q11 and I cannot find a cross for this? The first amp has 2SD2012 for Q11 can I just use this? The EBC pinout is different from the MPSU06 so I would need a few cuts/jumpers.
I cannot even find a datasheet for the MPSU06, the closest I could get is MPSU01. Could someone show me where the MPSU06 datasheet is?

thanks,

-jason

[chuck]

On Wed, 21 May 2014 13:21:35 -0700 (PDT), Jason Kooner

Jason,
http://www.abra-electronics.com/products/MPSU06-Transistor-Audio-NPN-2A-80V.html
Here is a link to the MPS U06.

The clipping issue depends on if the other side clips with a slight
increase of the input signal. If it doesn't, there is still a
problem. (Driver or output circuitry, something open) Chuck

[knd.k...@gmail.com]

I have the same speakers with same problem. How do you get to the amp? Get them open?