Nakamichi PA7II- are 2SA1837 and 2SC4793 suitable replacements for 2sc3298 and 2sa1306 in this amp?

Jump to Latest
#82
HighTec said:
Congrats ! So, the case can be closed ? 👍

Just one more tip. Do the bias adjusting only when the amp is cold.
Click to expand...
Its done thanks but did have question about what you said on bias.. On bias a tech years ago told me to start out at 0 when cold and slowly move up over a 30 minute period till I get to 40mv on this amp. Thats how I had it set but set like this when cold its around 70mv. Are you saying set to factory voltage when fully cold?
 
#83
Opinions may differ, but I believe HighTec's advice and your tech friend's advice arrive at the same result. The Nakamichi manual instructs to trim bias after 20 minutes or more. But as part of a repair session, the most cautious procedure would bring up the bias from minimum.

The fact that your amp starts high and settles lower indicates the temperature feedback is doing a good job. I think your amp is in a good place.

Enjoy!
 
  • Like
Reactions: HighTec
#84
Right said BSST. This is what for example QSC recommends when adjusting bias -->

4.1 Setting bias
Always set the bias
• after replacing any output or driver transistor.
• after replacing any diode or resistor in the driver/output circuitry.
• if the amplifier seems to run too hot at idle.
• if the amplifier exhibits crossover distortion.
The bias network sets the quiescent base current in the NPN and PNP driver transistors, which in turn sets the quiescent current in the output transistors. The driver transistors should both be slightly “on” at idle so that the transitions of the signal voltage between positive and negative are smooth and free of gaps or glitches. Too much bias current will cause the amplifier to run hotter than it should, especially at idle, while too little will cause noticeable crossover distortion, especially at low signal levels. The amplifier circuitry must be cool, or at least within a couple degrees of ambient air temperature, and the top cover must be removed. If the driver and output transistors are significantly warmer than the ambient air, leave the amplifier off and let it cool before proceeding. Before turning the amplifier on to set bias on one or both channels, familiarize yourself with the locations of the trimpots (R131 and R231) and the voltage measuring points so you can work quickly but thoroughly. If the amplifier warms up before you finish setting the bias, you will need to shut the amplifier off and let it cool down before you resume.
Click to expand...

Btw, below is the service manual for QSC RMX amplifiers.
 

Attachments

Last edited:
  • Like
Reactions: BSST
#87
Speaking of which. Does anyone have a service manual or schematic of Ecler's XPA11000 amp ? I politely asked Ecler for it but they told me it's a company secret. Secret ? For Pete's sake, this model is over 10 years old ! So no help from them. Acshually, I've been following Ecler's production timeline since the 90's when they launched their PAM (switching power mosfet) series with IRFP240 / IRFP9240 in the output. Long story short, I wanted to see what kind of engineering is involved in XPA11000. It's a class H, 3 tier design. According to available pictures on the internetz, it looks like they're using 12 pairs of IRF640 / IRF9640 in the output. I can see 4 small vertical mounted PCBs (step driver boards) with or probably it's a comparator LM311. Unfortunately there are no high resolution pics out there. I have service manual for XPA3000, 5000 and 7000 model but 11000 is not included although the specifications of this model is shown. Weird !

Sorry @gto127 for hijacking your thread. 😉

XPA3000, 5000 and 7000 is below if someone is interested. Cheers.
 

Attachments

#88
BSST said:
Opinions may differ, but I believe HighTec's advice and your tech friend's advice arrive at the same result. The Nakamichi manual instructs to trim bias after 20 minutes or more. But as part of a repair session, the most cautious procedure would bring up the bias from minimum.

The fact that your amp starts high and settles lower indicates the temperature feedback is doing a good job. I think your amp is in a good place.

Enjoy!
Click to expand...
I ended up doing bias by service manual. It was barely warm and only next to transistors. Just an update-DC offset is down to 7mv now. Only one thing bothers me at this point Q114 and 115 still run hotter than the other side-not scorching hot like before but hotter. Especially Q115. I was thinking mabye they got partially damaged when I had the thermal runaway due to bad connection. I have not been able to find new versions of these transistors. I found a new Q114 and a pull (used) Q115. Would it be ok to put a new and used in this case? They are from a reputable distributor. Also after I run amp for 2 hours the volume seems to lower a little not sure if this is normal for these amps or not.
 
#89
I doubt the transistors got damaged by thermal stress, but just IMHO. Comparing voltage across R117 with similar voltages across R116 (and other channel) will tell you how similar bias currents are re other analogous transistors. Another measurement that might be insightful is the voltage between emitter of Q114 and emitter of Q115, compared with other channel. Significant differences might point to some remaining damage in output transistor array. I do think it's safe to swap transistors if you remain suspicious.

I'm happy to opine if you find something concerning.

I don't know why gain would drop with time or temperature. Can you confirm this impression with test equipment?

Good luck,

Steve
 
  • Like
Reactions: HighTec
#90
BSST said:
I doubt the transistors got damaged by thermal stress, but just IMHO. Comparing voltage across R117 with similar voltages across R116 (and other channel) will tell you how similar bias currents are re other analogous transistors. Another measurement that might be insightful is the voltage between emitter of Q114 and emitter of Q115, compared with other channel. Significant differences might point to some remaining damage in output transistor array. I do think it's safe to swap transistors if you remain suspicious.

I'm happy to opine if you find something concerning.

I don't know why gain would drop with time or temperature. Can you confirm this impression with test equipment?

Good luck,

Steve
Click to expand...
I do not have a amp power meter but did notice the amperage draw on front of my surge suppressor lowered amperage draw as time went on. Think it started around 4 amps and went down to around 1.5. I've still got it together at moment. Is any risk of running like this for time being?The previous bad side of amp has about same heat on heatsink as other side with exception of Q114/115. If Q115 did go bad would it take out other transistors or just Q115?
 
#91
HighTec said:
Speaking of which. Does anyone have a service manual or schematic of Ecler's XPA11000 amp ? I politely asked Ecler for it but they told me it's a company secret. Secret ? For Pete's sake, this model is over 10 years old ! So no help from them. Acshually, I've been following Ecler's production timeline since the 90's when they launched their PAM (switching power mosfet) series with IRFP240 / IRFP9240 in the output. Long story short, I wanted to see what kind of engineering is involved in XPA11000. It's a class H, 3 tier design. According to available pictures on the internetz, it looks like they're using 12 pairs of IRF640 / IRF9640 in the output. I can see 4 small vertical mounted PCBs (step driver boards) with or probably it's a comparator LM311. Unfortunately there are no high resolution pics out there. I have service manual for XPA3000, 5000 and 7000 model but 11000 is not included although the specifications of this model is shown. Weird !

Sorry @gto127 for hijacking your thread. 😉

XPA3000, 5000 and 7000 is below if someone is interested. Cheers.
Click to expand...
"Sorry@ @gto127 for hijacking your thread." Thats certainly alright-anything that helps.
 
#92
gto127 said:
I do not have a amp power meter but did notice the amperage draw on front of my surge suppressor lowered amperage draw as time went on. Think it started around 4 amps and went down to around 1.5. I've still got it together at moment. Is any risk of running like this for time being?The previous bad side of amp has about same heat on heatsink as other side with exception of Q114/115. If Q115 did go bad would it take out other transistors or just Q115?
Click to expand...
I don't know if Q115 (or others) failure would lead to a catastrophic cascade. The amp does feature fusing.

But no doubt you are familiar with Murphy's law. I remember one of its corollaries: "A transistor protected by a fast-acting fuse will protect the fuse by blowing first." 😢
 
#93
BSST said:
I don't know if Q115 (or others) failure would lead to a catastrophic cascade. The amp does feature fusing.

But no doubt you are familiar with Murphy's law. I remember one of its corollaries: "A transistor protected by a fast-acting fuse will protect the fuse by blowing first." 😢
Click to expand...
HaHa- I agree with the fuses. They are there to make the consumer feel good.
 
#95
BSST said:
Comparing voltage across R117 with similar voltages across R116 (and other channel) will tell you how similar bias currents are re other analogous transistors. Another measurement that might be insightful is the voltage between emitter of Q114 and emitter of Q115, compared with other channel. Significant differences might point to some remaining damage in output transistor array.
Click to expand...

+1

Btw, your amp does not have volume control pots, clamp the inputs to the ground before measuring. Just in case there is no noise picked up by the open inputs.

Edit: well hello there wg_ski ! Long time no see man ! 🙂
 
#98
BSST said:
I don't know if Q115 (or others) failure would lead to a catastrophic cascade. The amp does feature fusing.

But no doubt you are familiar with Murphy's law. I remember one of its corollaries: "A transistor protected by a fast-acting fuse will protect the fuse by blowing first." 😢
Click to expand...
Almost had a bad failure. Got the used 2sa1306b and 2sc3298 b's in. Powered it up the lights went dim and bias went to over a volt. Unplugged immediately. Put back in the old ones and thank God everything was ok. Dont think I will ever buy a used electronic part again. These parts new are just unobtainable it seems. I think I can find the a versions which are 180 volt instead of 200. Would these work reliably?
 
#100
Why would you change 2SA1306/2SC3298 if they're working OK ? :scratch2:

Edit: btw, the only substitutes I came across are 2SA1837/2SC4793 still manufactured by UTC. There is also 2SA1668/2SC4382 manufactured by SanKen.

https://www.profusionplc.com/type/transistor
 

Attachments

Last edited:
You must log in or register to reply here.
Top Bottom