Welcome reader,
I'm looking for replacement caps for my 2 TPA3116 2.1 50Wx2+100W Amplifier Board's as one 1 board the big cap's head is bulging out.
The current cap reads Nichicon 6800uF 35V VZ(M) and is fake because it's to small to be real. It measures about 37x17mm. (real 40x20mm).
It would be an aluminium electrolytic cap for wide temperature range.
Putting in a real one will be a sqeeze, but putting in 2 in vertical position is also possible as the board has holes for them and would be an advantage over 1 big one.
After reading some bit about caps I think low esr cap's would be best to use, is this true?
As a replacement I've found these 2pcs 4200uF 35V JAMICON WG 18X45mm 35V4200uF Low ESR Long Life Capacitor 1 or 2 will fit my board if mounted vertically.
Any better options?
Some advice please!
I'm looking for replacement caps for my 2 TPA3116 2.1 50Wx2+100W Amplifier Board's as one 1 board the big cap's head is bulging out.
The current cap reads Nichicon 6800uF 35V VZ(M) and is fake because it's to small to be real. It measures about 37x17mm. (real 40x20mm).
It would be an aluminium electrolytic cap for wide temperature range.
Putting in a real one will be a sqeeze, but putting in 2 in vertical position is also possible as the board has holes for them and would be an advantage over 1 big one.
After reading some bit about caps I think low esr cap's would be best to use, is this true?
As a replacement I've found these 2pcs 4200uF 35V JAMICON WG 18X45mm 35V4200uF Low ESR Long Life Capacitor 1 or 2 will fit my board if mounted vertically.
Any better options?
Some advice please!
I agree with you, two in parallel offer better ESR than only one (same type) because also the ESR will be in parallel. This is a class D amplifier with a high frequency carrier frequency and accordingly a high frequency component in the current consumption. As I am not sure other (smaller) decoupling capacitors are included to take up the high frequency current ripple, I would also use low ESR types.
The ones you suggest are probably fine.
A small remark: I believe electrolytic capacitors are frequently specified with maximum dimension such that they may be smaller if the production technology allows a more compact outline. Evidently, the lead distance has to remain correct. So fake or not, difficult to say. But, the bulging is no good sign.
The ones you suggest are probably fine.
A small remark: I believe electrolytic capacitors are frequently specified with maximum dimension such that they may be smaller if the production technology allows a more compact outline. Evidently, the lead distance has to remain correct. So fake or not, difficult to say. But, the bulging is no good sign.
Last edited:
Thank you.
I suspect the current 6800uF cap to be one like this ($1), but it's black with white and says Nichicon. The rest of the text, looks and size is the same.
The current cap is rated @35v and I have used it at max 24V for a few months now, so it should not have died. Fake or not, the replacement will be 2 low esr/impedence caps.
Is there much difference in brands?
I allready linked the Jamicon 2x 4200uF 2 for $4
3300uf rubicon 3 for $6
3000uF ELNA 6 for $5 and the same ones in 5600uF or 6500uF are 2 for $6
The Elna's have the best size; would I get an $4 worth of improvement by placing 2 caps with higher capacitance vs 2x 3000uF? (3 times the price)
Lead distance (length) will be a bit shorter with 1 or 2 caps vertically vs 1 (bigger) placed horizontally as the leads stick out of the board for about 10mm.
I suspect the current 6800uF cap to be one like this ($1), but it's black with white and says Nichicon. The rest of the text, looks and size is the same.
The current cap is rated @35v and I have used it at max 24V for a few months now, so it should not have died. Fake or not, the replacement will be 2 low esr/impedence caps.
Is there much difference in brands?
I allready linked the Jamicon 2x 4200uF 2 for $4
3300uf rubicon 3 for $6
3000uF ELNA 6 for $5 and the same ones in 5600uF or 6500uF are 2 for $6
The Elna's have the best size; would I get an $4 worth of improvement by placing 2 caps with higher capacitance vs 2x 3000uF? (3 times the price)
Lead distance (length) will be a bit shorter with 1 or 2 caps vertically vs 1 (bigger) placed horizontally as the leads stick out of the board for about 10mm.
The capacitor you show from Ebay is what I typically use and have good experience with. On a forum I read that a lot is produced under the name "Nichicon" but not all are equally good. Perhaps, there are more manufacturers behind the name Nichicon. Low ESR is a subjective parameter. Anyway, your capacitor should not have died after short use at 24V. You better change them both.
I once bought low ESR capacitors from another Asian brand. 25V rating. Short time of use at 24V made two explode. Two means, at least that production batch was flawed. OK, the capacitors were cheap and I used them for 12V purposes where they seem to work fine.
If you buy capacitors from Elna or Philips you are much more certain that they will stand the specs. Asia can produce good capacitors, you are just not sure which ones are good. It is a matter of brand control and price.
Which to choose for your present amplifiers? My experience with Asian capacitors is most of them work to my satisfaction. You need not use Elna all the time. For power supplies I build to be used with different amplifiers in the future, I sometimes use Elna to have less doubt about their performance. For amplifiers I make as experiments and use for up to a couple of years, I am not prepared to pay Elna prices. The amplifier was cheap to make as another experiment and why make it more expensive with Elna capacitors? I do not believe you can unambigously hear if the decoupling capacitors are Elna or not.
I once bought low ESR capacitors from another Asian brand. 25V rating. Short time of use at 24V made two explode. Two means, at least that production batch was flawed. OK, the capacitors were cheap and I used them for 12V purposes where they seem to work fine.
If you buy capacitors from Elna or Philips you are much more certain that they will stand the specs. Asia can produce good capacitors, you are just not sure which ones are good. It is a matter of brand control and price.
Which to choose for your present amplifiers? My experience with Asian capacitors is most of them work to my satisfaction. You need not use Elna all the time. For power supplies I build to be used with different amplifiers in the future, I sometimes use Elna to have less doubt about their performance. For amplifiers I make as experiments and use for up to a couple of years, I am not prepared to pay Elna prices. The amplifier was cheap to make as another experiment and why make it more expensive with Elna capacitors? I do not believe you can unambigously hear if the decoupling capacitors are Elna or not.
Last edited:
Which one of the 3?
Most important for me is that they work longer then the current one
Prices to replace them per board are:
1x noname 6800uF for $1.15 or 2 smaller ones are about the same price.
2x Elna 3000uF for $1.70
2x 4200uF Jamicon for $4
2x Elna 5600/6500uF for $6
As I really like these amps and use them almost daily, they are well worth the mayor investment in some brand caps.
My guestimate is that the Elna 2x3000uF already are a big upgrade and will be fine but how much better are bigger ones and is bigger really better/noticeable here? Can they be to big and have bigger caps downsides other then space and cost?
I looked in the tpa3116 specsheets but there is no sizing info for these caps. My other tpa3116 (single chip) boards have single Rubicon 25V 3300uF or single United Chemi-Con 35V 1000uF caps. The latter one is probably a bit small and can get an upgrade as well.
The ones I often use are the Nichicon green-gold type. No bad experiences. I believe when I earlier followed your link on the word "fake", I got to an Ebay advertisement for a green-gold Nichicon 6800uF 35V. Now I only get an error with that link.
2x3000uF Elna should be fine.
I am no expert in electrolytic capacitor technology. My impression is that the size depend on capacity/voltage, production technology and voltage margins in the design. I would assume that bigger Elna capacitors are at least no worse electrically than smaller Elna capacitors with the same capacity and voltage rating. The bigger ones may even have better voltage margins.
Is an electrolytic capacitor specialist available?
2x3000uF Elna should be fine.
I am no expert in electrolytic capacitor technology. My impression is that the size depend on capacity/voltage, production technology and voltage margins in the design. I would assume that bigger Elna capacitors are at least no worse electrically than smaller Elna capacitors with the same capacity and voltage rating. The bigger ones may even have better voltage margins.
Is an electrolytic capacitor specialist available?
That link is indeed incorrect
; this should work http://www.nichicon-us.com/english/products/pdf/e-vz.pdfThis looks just like the swollen/bulging one on the board.
Where did you find the measurements?
Strangly I can't even find them at that webshop (so good find!) searching for Nichicon and the price is only twice of that of a similar one sold as Nichicon at 1/13th (470uF) capacity. Is this normal / to be expected, or is there some Nichicon fake or economy/consumer/die quickly product line?
This cap is with it's 40 cents even cheaper then the cheapest I could find on ebay; to bad shipping is €5/$6.
The horizontally mounted cap does not seem to be a very good idea because of the longer leads and ability to only use 1 cap, but when this caps head swells, it pushes the leads towards the heatsink and can short isself out...
Thx, so those at Polin measure the specified 40x20mm instead of the one on my board which looks exactly the same but is 37x18mm (measured again in proper light)
I found a cheaper seller in the UK for the elna 5600uF but he only had 2 in stock and has more in 3 weeks. 6 of those with shipping would cost about $13/€12
I ordered the 6x Elna 3000uF for $4.99 / €4.32 incl. airmail.
Thank you and especially FauxFrench.
I found a cheaper seller in the UK for the elna 5600uF but he only had 2 in stock and has more in 3 weeks. 6 of those with shipping would cost about $13/€12
I ordered the 6x Elna 3000uF for $4.99 / €4.32 incl. airmail.
Thank you and especially FauxFrench.
They arrived 2 days ago and all measure 3810-3830uF instead of the 3000uF specified. They fit the boards perfectly and seem to work fine. The bulged cap measured only 1500uF and the other removed one from the other board 6500uF like it specs.
Time to make some nice enclosures for them and a few other boards. I could not find much cases for inspiration. Is there a 'show your enclosure' topic on this forum or somewhere else on the web?
Time to make some nice enclosures for them and a few other boards. I could not find much cases for inspiration. Is there a 'show your enclosure' topic on this forum or somewhere else on the web?
Last edited:
If price doesn't matter, these caps are the newest latest in performance, Chemi-Con GPD:
http://www.chemi-con.com/upload/files/7/1/134396053452d6c274f0cce.pdf
These will last forever as they're speced for 3000h at 125/135°C at full voltage and ripple.
Every -10°C will double their lifetime.
115°C: 6000h
105°C: 12000h
95°C: 24000h
85°C. 48000h
75°C: 96000h
still at full voltage and ripple.
http://www.chemi-con.com/upload/files/7/1/134396053452d6c274f0cce.pdf
These will last forever as they're speced for 3000h at 125/135°C at full voltage and ripple.
Every -10°C will double their lifetime.
115°C: 6000h
105°C: 12000h
95°C: 24000h
85°C. 48000h
75°C: 96000h
still at full voltage and ripple.
How does this relate to other caps?
The 'cheap' Elna RJD's I put in the boards are 'only' specified at 8000hours@105°C
They measure 20% higher then specified, so they should still perform at the specified 3000uF after 8000hours@105°C
And as it is not the power but heat which kills electronics; I'm thinking of putting in a small 24V cage fan and connect it to the dc input or use a 12V version and a fan controller instead of making passively cooled cases. Maybe with just an on/off switch for when it's hot or when I use it at high power/voltage or in a warm environment.
Any enclosure / case idea's are very welcome.
My idea about a case is very simple: triplex bottom and top, multiplex sides so it will show all the layers and a piece of metal from an old frying pan for the front and back. I could line the inside with metal as well and/or make the bottom and back out of the same piece of metal sheet.
Some holes (with a piece of mesh?) for air flow.
Size about 14cm x 10cm deep or more if I make 'ears' on the back and/or front) x 4,4cm high. (1/3 the width of a standard rack unit and 1u high.
This normally relate to all wet/polymer caps from almost all vendors. Some use slightly different formulars to predict more precise.
-10°C -> double the lifetime
/2 ripple current -> double the lifetime
/2 applied voltage amplitute -> double the lifetime
For other caps like foil, ceramic, tantal this isn't valid.
-10°C -> double the lifetime
/2 ripple current -> double the lifetime
/2 applied voltage amplitute -> double the lifetime
For other caps like foil, ceramic, tantal this isn't valid.
Reliability is a very relevant factor for circuits that may endanger persons or property if failing.
You are two very serious engineers so you also take that parameter into account. When I was very young, I made electronic circuits without thinking about lifetime and when I found them by chance 40 years later they still worked.
My recommendation to DIYs is to make circuits that are reliable but not to worry about lifetime. In three years time your constructions are either taken apart or in the drawer.
You are two very serious engineers so you also take that parameter into account. When I was very young, I made electronic circuits without thinking about lifetime and when I found them by chance 40 years later they still worked.
My recommendation to DIYs is to make circuits that are reliable but not to worry about lifetime. In three years time your constructions are either taken apart or in the drawer.
True, but there is also some personal pride in that my diy builds or repairs hardly fail. Nowadays lots of consumer products look like the are engineered to fail after warranty.
I like these amps very much and use 1 as my main amp with a test setup of drivers with 4 Visaton frs 8m 3.3" in pvc pipes. 2 On each side in a 120degree angle and a Dayton PTmini-6 planer tweeter in the middle.
I Still have to build the final speaker boxes for them and use them with 2 jbl804 closed car subs; but I am amazed at how much I love the sound from this simple and cheap setup. I'm actually a bit worried that the final box is going to sound less good, so I'm looking more into why it's sound so good at the moment.
As I'm writing and reading this, this might just be a lame excuse for not having them build yet....
I like these amps very much and use 1 as my main amp with a test setup of drivers with 4 Visaton frs 8m 3.3" in pvc pipes. 2 On each side in a 120degree angle and a Dayton PTmini-6 planer tweeter in the middle.
I Still have to build the final speaker boxes for them and use them with 2 jbl804 closed car subs; but I am amazed at how much I love the sound from this simple and cheap setup. I'm actually a bit worried that the final box is going to sound less good, so I'm looking more into why it's sound so good at the moment.
As I'm writing and reading this, this might just be a lame excuse for not having them build yet....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.