reduce power to reduce heat from UPC1342V kit

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
If it gets so hot that it smokes, the transistors are probably ruined. Ruined may mean short-circuited. If short-circuited, the power resistors may have much more voltage across them and they burn.
With a modest price and unless you are experienced with amplifier fault finding, you better buy a new module. Else you have to find and order replacement parts, await arrival etc....

Yeah, that's exactly what I'll do, the kit without caps was about $10, so its certainly not worth the time to troubleshoot and replace individual parts.

I'm on vacation from Monday, so amplifier fun has to be put on hold, but thanks to everyone who has helped a noob like me get my system sorted.
 
Small update, after getting stuck in Japan for almost 5 weeks, I am back at my multi-meter finishing off the project.

Built another amp board after binning the old one, and we're currently functional and sounding pretty great!

Currently got the mV set to around 50mV across those 5W resistor and the output volume is a little more than enough for what I need but the heat created is still a little too much for my liking (ie, the wife is gonna complain).

So, 2 options are to fit a fan, which I have allowed for in the case layout and I have a spare 12v source, so thats no big problem, or, reduce the mV trim pot a little more.

Reducing the voltage, with the current status of the case, it would be a big pain in the *** to access the trim pot and then test, so I really only wanna do that if its going to make a significant difference.
 
ran the amp for half a day on and off, and botched a 120mm fan in the top as a test and my conclusion of if I need a fan OR to lower the mV: YES!

Gonna have to do both, I'd wake up in a cold sweat worrying if i remembered to turn off the amp.

Ordered a fan cover, so once that's here I'll set about cutting a hole in the lid and lowering the voltage a little more, then hopefully she'll be ready for her debut in the amp gallery on here.
 
Hi nine0nine


yes lowering the supply voltage (as nigel wrote) will help for sure. you can use a fan but try to find a fan with low noise like some 17dB or 18db fans like the PC gamer us it. normally these fans are using PWM rpm- so look at some "normal" 12V fans with low noise.
you can also re check if you have enough ventilation through your amp housing
chris
 
Hi nine0nine


yes lowering the supply voltage (as nigel wrote) will help for sure. you can use a fan but try to find a fan with low noise like some 17dB or 18db fans like the PC gamer us it. normally these fans are using PWM rpm- so look at some "normal" 12V fans with low noise.
you can also re check if you have enough ventilation through your amp housing
chris

Thanks for the fan info.

I have a few fans here I was using for testing, one was silent, but didn't move much air, the other blew a gale but was pretty loud, gonna see if i can run it at a lower voltage to find a compromise.

Currently waiting on a few bits to be delivered before I tweak it, but running it for a few hours today was bliss after all the problems I've had :D
pic, if anyone is curious...
amp1_zpsxieyexwm.jpg
 
Hi
the housing seems to be very compact = less space but you got it! very good.
if i look inside i can guesstimate that the bottom plate has no drilled holes and you get no air convection through the amp. how big are your feets for that amp? its better to drill under the amp boards some holes( as much as better) and let the air flow + have higher feet the start free air convection. thermal past at the chips and heat sink and if you want connect the front plate and the heat sink with thermal paste too.


chris
 
Last edited:
Thanks for the fan info.

I have a few fans here I was using for testing, one was silent, but didn't move much air, the other blew a gale but was pretty loud, gonna see if i can run it at a lower voltage to find a compromise.

Currently waiting on a few bits to be delivered before I tweak it, but running it for a few hours today was bliss after all the problems I've had :D
pic, if anyone is curious...


Is it still running well?


May I ask what supply voltages and speaker impedance you are using?
 
Does anyone have any experience or wise suggestions with regards to lowering the uPC1342V gain and selecting the compensation capacitor?

I currently have 1.8k and 56k for gain (see attached schematic) and 22pF. (Schematic shows 15pF, board came with 22pF.)

I would like to use 5.6k and 56k for gain. However I would like to appropriately adjust the compensation capacitor at the same time.
 

Attachments

  • Datasheet Page 3.png
    Datasheet Page 3.png
    74 KB · Views: 131
Last edited:
There may be "the lowest available gain of upc1342" in the manual
It may not work properly at 11x magnification. Friends, you have to be careful.:)

I think the 12-22pf capacitor is suitable. It can adjust SR according to specific needs. There may be these contents in the manual. Maybe you can try more complicated compensation methods, such as double pole compensation.
 
There may be "the lowest available gain of upc1342" in the manual
It may not work properly at 11x magnification. Friends, you have to be careful.:)

I think the 12-22pf capacitor is suitable. It can adjust SR according to specific needs. There may be these contents in the manual. Maybe you can try more complicated compensation methods, such as double pole compensation.


I will look at the datasheet again. I asked the question because I am concerned that it might not be stable if I lower it. I was hoping someone else had lowered the gain successfully. If not I might need to leave it alone.
 
There may be "the lowest available gain of upc1342" in the manual
It may not work properly at 11x magnification. Friends, you have to be careful.:)

I think the 12-22pf capacitor is suitable. It can adjust SR according to specific needs. There may be these contents in the manual. Maybe you can try more complicated compensation methods, such as double pole compensation.

Unfortunately the datasheet I have does not give information on compensation or gain options.

Perhaps I will try all my other experiments first and leave the gain and compensation as a last experiment. (With a DBT and small fast fuse.)

I have adjusted the gain of amplifiers like the MX50SE successfully but this driver IC is a little bit more of a black box. I have attached the equivalent circuit. The compensation capacitor is connected to pins 6 and 7.
 

Attachments

  • Datasheet Page 6.png
    Datasheet Page 6.png
    76.4 KB · Views: 67
  • Datasheet Page 5.png
    Datasheet Page 5.png
    85.1 KB · Views: 67
  • Datasheet Page 4.png
    Datasheet Page 4.png
    86.7 KB · Views: 102
  • Datasheet Page 3.png
    Datasheet Page 3.png
    74 KB · Views: 98
  • Datasheet Page 2.png
    Datasheet Page 2.png
    133.8 KB · Views: 106
  • Datasheet Page 1.png
    Datasheet Page 1.png
    173.2 KB · Views: 111
  • Equivalent Circuit.png
    Equivalent Circuit.png
    37.7 KB · Views: 124
Unfortunately, there is nothing in the manual that we need.
The capacitors between pins 6 to 7 are Maitreya capacitors. Refer to LM4702, LME49810 and other similar ICs, and their value may be between 10-25pf. When the gain is 30 times, we can try to input a 20kHz square wave, and then gradually reduce the 15pf capacitor. If the rising edge of the square wave begins to overshoot, the capacitance value at this time is the smallest usable value. For safety, we can appropriately increase the minimum usable value of the Maitreya capacitor in actual use.

The manual does not provide the minimum usable gain or open-loop phase frequency diagram. But refer to the typical examples in the manual, and the situation of similar ICs such as LM4702 and LME49810. Similarly, we can gradually reduce the gain and observe whether the amplifier oscillates, and then leave a proper safe space. I think using 30x magnification is safe and applicable.

The smaller the resistance, the lower the noise. If you use a magnification of about 30 times, I suggest using 500R and 14k resistors. At the same time, you must also adjust the value of the coupling capacitor and the negative feedback capacitor. Capacitors are not expensive, and it is not difficult for modern audio sources to drive 10k resistance loads, so I think these costs are worthwhile.

After all the above are done, we can try to add more complicated compensation methods such as double pole compensation to further reduce the distortion.

Looking forward to your good news.:p
 
The input low pass filter and the compensation capacitors appear to be tiny ceramic capacitors. I don't know the manufacturer, dielectric type or rated voltage. (Picture enclosed.) Hopefully they are NP0 but I don't know what the board was built with.

Does anyone have any comments on suitable types of capacitors for both purposes?

I was thinking of putting a small (10-15 pF) cap across the 56k feedback resistor and/or increasing the compensation capacitor prior to experimenting with reducing the gain.


In the future I would like to add a real oscilloscope to my bench. Right now I have a couple (slow) home made oscilloscopes (STM and Arduino based) and the common DSO138.
 

Attachments

  • LPF and Compensation Capacitors.jpg
    LPF and Compensation Capacitors.jpg
    393.9 KB · Views: 49
Yes, the reason I am asking is because I am concerned that it will become unstable.

Perhaps I will only go as far as 26 dB and see what difference that makes. From there perhaps I will increase the compensation cap and/or try a small cap across the feedback resistor to see if that helps at all. Perhaps that will be the limit for this board until I have more suitable equipment.

Right now I am listening to how the changes affect impulsive percussion sounds like splash cymbals. When I get a better scope at home I can look at 20 kHz square waves.


Perhaps I can make use of REW to make quantitative evaluations of the changes?
 
I have attached a photo of my uPC1342V where I used an external rectifier and filter board to make sure that the rectifier and noise/interference from the rectification of the large signal AC power would not be a problem (so close to the input components/stage).

I did my first distortion measurements up to 8Vrms into a 4Ohm test load. The load is getting pretty hot so I might get a larger one before going higher.

I would appreciate expert comments and interpretation of the distortion measurements. (Keeping in mind it is a 4Ohm load not 8Ohm, and keeping in mind that I only went up to 8Vrms.)
 

Attachments

  • Second Style uPC1342V Kit First Test.jpg
    Second Style uPC1342V Kit First Test.jpg
    997.5 KB · Views: 84
  • 1kHz Spectrum 2nd Design uPC1342V 4Ohms 8VRMS Upper Terminals Side Far From Transformer.png
    1kHz Spectrum 2nd Design uPC1342V 4Ohms 8VRMS Upper Terminals Side Far From Transformer.png
    141.6 KB · Views: 80
  • uPC1298V Datasheet 4 Ohm Harmonic Distortion.png
    uPC1298V Datasheet 4 Ohm Harmonic Distortion.png
    89.9 KB · Views: 86
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.