PPY,
Some questions I would ask are what is the efficiency of your speakers and how loud do you want them to go? This would answer the question of the power needed. Another question is what the frequency response and impedance of your speakers is, what is the f3 tuning of your box. Then I would want to know what the FR of the amplifier is. Without this information and more we don't have any real idea of what you are complaining about with the bass output.
Some questions I would ask are what is the efficiency of your speakers and how loud do you want them to go? This would answer the question of the power needed. Another question is what the frequency response and impedance of your speakers is, what is the f3 tuning of your box. Then I would want to know what the FR of the amplifier is. Without this information and more we don't have any real idea of what you are complaining about with the bass output.
I really have to give these guys credit for being open with their design, answering questions, taking suggestions, and taking measurements on it with test equipment for us. Not many manufacturers will do that.Well I think the credit factor is not that big with chinese boards!
Well I think the credit factor is not that big with chinese boards!
That generic comment is quite useless and inappropriate IMHO
Oh, I meant crest factor.
I suspect the primary bulk cap to be bottleneck, as it is only 200uF if I remember correctly? So at higher output, secondary voltage sags which results in higher switching current within the amp thus then trips the internal over current-protection. I know this behavior from when using a current-limited boost supply (like these cheap 10-15A simple switchers based on TL494 or UC38XX) with these amps, it's pretty normal/common. But before the OC-protection kicks in, sagging voltage phenomenon is indicated by the clipping indicator.
I suspect the primary bulk cap to be bottleneck, as it is only 200uF if I remember correctly? So at higher output, secondary voltage sags which results in higher switching current within the amp thus then trips the internal over current-protection. I know this behavior from when using a current-limited boost supply (like these cheap 10-15A simple switchers based on TL494 or UC38XX) with these amps, it's pretty normal/common. But before the OC-protection kicks in, sagging voltage phenomenon is indicated by the clipping indicator.
Last edited:
tpa3251 G3H module reply！
hello, everyone!
We go to work, stayed testing yesterday, continues today to answer questions, welcome to all kinds of questions about the products.
I have more free in this week, will very busy next week.
Answer some questions:
About SMPS:
1) this SMPS topic is qasi resonant FLB , better efficiency lower EMI.
2)Transformer can bear 300W rms power with this circuit under good cooling system, if run continous , please add one fan cooling aside.
3)About AC input capacitor,if AC supply =220v -240v, 0.8uF-1uF/1w, 220uF-470uF/400V electric capacitor should satisfy with this board.
If AC supply =100v -120v, 2.4uF-3uF/1w, 680uF-1000uF/400V electric capacitor should satisfy with this board.
we select 220uf/400v, it is conservative value, bigger capacitor need more money, recommand to increase this cap in lower ac supply area,
4) This SMPS have good protection, support over temperature shutdown and over current shutdown protection, this protection is latched protection, it can not recovery until you power off the AC input.
5)Usually , SMPS and class D amp 1:1 power supply should no propblems, so we leave no more room here.
6) According to EIAJ measurment standard, this design should no problems,friends here want this module run like giant， and cost must be dwarf。
hello, everyone!
We go to work, stayed testing yesterday, continues today to answer questions, welcome to all kinds of questions about the products.
I have more free in this week, will very busy next week.
Answer some questions:
About SMPS:
1) this SMPS topic is qasi resonant FLB , better efficiency lower EMI.
2)Transformer can bear 300W rms power with this circuit under good cooling system, if run continous , please add one fan cooling aside.
3)About AC input capacitor,if AC supply =220v -240v, 0.8uF-1uF/1w, 220uF-470uF/400V electric capacitor should satisfy with this board.
If AC supply =100v -120v, 2.4uF-3uF/1w, 680uF-1000uF/400V electric capacitor should satisfy with this board.
we select 220uf/400v, it is conservative value, bigger capacitor need more money, recommand to increase this cap in lower ac supply area,
4) This SMPS have good protection, support over temperature shutdown and over current shutdown protection, this protection is latched protection, it can not recovery until you power off the AC input.
5)Usually , SMPS and class D amp 1:1 power supply should no propblems, so we leave no more room here.
6) According to EIAJ measurment standard, this design should no problems,friends here want this module run like giant， and cost must be dwarf。
About class D stage,
1)TPA3251 have very good sound performance in the class D amplifier area.
2)All the class D amplifier have weak speaker control ability compare to class AB or class A or B, so we feel weak bass output.
but we can add prefilters or DSP unit to improve it, but we have to admit it is different bass from two kinds of amplifer delivery out.
It is common consensus in this area.
3) This amp support one by one channel testing, not support double channels high load same time, maybe chip small and temperature protection occur, so we find blink when stereo high load, but alone channel testing no problems, it can delivery out high power.
4)if find led blink when testing, it is chip protection, not power supply problems, power supply no protection under this condictions.
About AMP parts
1)All the SMT ceramic capacitors come from TDK factory in China,performance well ,even batch applied in our high quality military products。
2）Inductors is ourself manufacture,Iron Silicon aluminum powder cores， it is our customerized order from China factory,only this factory have best performance in this area, we have compare to many factory cores (inlude American Magenatic) and many type cores, this cores performance well , lower fever，high DC resistance of saturated capacity ，High product consistency。
we have retest in the tpa3251 EVM board, better than EVM board inductors indeed。
BTW, lower price mean high produce ability, not only Chinese make lower level lowest price products, but also make high level high price/performance products.
hello ,ppy, about DC voltage, we use seperate SMPS with EVM board test again this morning,lucky we found main DC voltae 30V works well, because we have 30V/600W SMPS aside, we use it drive EVM board, test stereo mode works well, not like 36Vdc ,
so we change R17 value to 1K8， main dc is 30V, test G3H with stereo mode, no blink again, it same all ok now.
I think TPA3251 easy protection under 36V main dc conditions，add lower voltage should okay.
I recommand to change the R17 to 1K8.
THANKS!
Attachments
Last edited:
TPA3251 evm BOARD SQUARE WAVE PHOTO
we usse seprate SMPS this morning, and drive the EVM board, add 50Hz square wave, load is 4ohm resister,main dc 30v
output wave below, can you clear all the details to us about this EVM circuit?
TKS!
Great!
we usse seprate SMPS this morning, and drive the EVM board, add 50Hz square wave, load is 4ohm resister,main dc 30v
output wave below, can you clear all the details to us about this EVM circuit?
TKS!
Attachments
Last edited:
Hello，Manfred
we change R17 value to 1K8， main dc is 30V, test G3H with stereo mode, no blink again.
I think TPA3251 easy protection under 36V main dc conditions，add lower voltage should okay.
I recommand to change the R17 to 1K8.
About transformer temperature, most of case is shielding cooper tape high temperature, the tape functions is sheilding the
magnetic flux leakage，EC core tempearture increasing very slow.it should no problems for this amp apply.
Last edited:
R17 resister position, below the transformer, red block.
Attachments
Revision documnet below, please download and revise the module, sorry for the inconvenience.
Attachments
That's strange, as the chip is proven to drive full continuous output power into 4R in stereo. I test it several times with a 38V supply for infinite time. 2x140W, no problems whatsoever..
With 30V supply you won't reach the stated 2x140W anymore. (Estimated 95W at 1% with 30V)
Any plots/measurements to prove this?
Could you answer me the question, why your Nichicon MP are green/gold rather than brown/white?
I m asking because a lot counterfeit Nichicons are green/gold, even if their are faked brown/whites too.
Last edited:
Correction:
I'm asking because a lot of counterfeit Nichicons are green/gold, even if there are faked brown/whites too.
hello, doctormord, please upload your test photos, maybe someone find the reason.
About indictor, we repalce the original with our indictor , and add load to run, we will retest it again and upload some photos tomorrow.
we have to go home now, our colleagues can not stay office too later today.
About capacitor, it is our customerized type, our factory order over 100thounds this capacitor a month, it is common type.we can order any color we need,maybe different of catalog list,but looks good than orginal.
Friends here find any other questions, let's discuss here, chinese saying: good good study, day day up.
Thanks！
Last edited:
