The fact that you can use the authentic IC in a socket type prototype board, get reasonably decent performance and short the output without there being oscillation is a testimony to how good the real part is
Thanks for making this video. I have only recently realized how widespread the problem is. I don't think most people who by parts on eBay have any idea what they are actually getting. After all the chip worked and how many hobbyist will actually measure output power ? They think they got a great deal on an authentic chip. These things are almost impossible to spot if you don't have an original for comparison or a video like yours.
i have bought plenty of counterfeit Amp IC's knowing good and well they were fake and as long as you build a quality circuit around it they work great, Just steer clear of the ones that don't show a real picture in the listing.
11:18 - Did you check your power-supply voltage to make sure it isn't changing when you attached the counterfeit part? Max power output is determined DIRECTLY from the rail-to-rail voltage!
Yes you are right. I didn't check that time but have in the past. The authentic ICs pull the supply voltage down more because they made more output power which draws more current off the supply. If tested on the regulated supply, there would be an even wider difference between the counterfeit and authentic parts.
Also be careful when dealing with lm1875s, a lot of tda2030 just marked as an lm1875. A couple of things i noticed are that 1875s dont have those side notches and have wider leads that are rounded at the end and not sharp as these tda chips. Those tdas that they engrave as lm1875 are probably fake as well. Great video as always John.
I was impressed by the counterfeit TDA2040 design. So I took some time to look at your picture in detail. The counterfeit die looks completely re-engineered. They have changed the size of the elements, they changed the structures, they put structures away. I can imagine, that this is a whole lot of effort. Why do they do this? Isn't it much easier and less expensive to keep the structures as they are? Maybe, the original structures don't fit in the manufacturer's technology, but I cannot imagine. These old devices are like pieces of art. The structures are big and wide and in each case you have to use special masks etc.
Isn't the likely explanation that the part is actually a lower power amp chip (possibly an out-of-spec one at that) that were bought cheaply (or obtained) from the actual manufacturer and re-labelled as TDA2040?
I live in India and here I don't find authentic tda2030A or 2050 in local shops and I also don't know whether these chips available on Amazon or any other online stores are genuine or not. Sometimes I do find a chip with wider legs, having that small round dot and also those two cuts positioned below as you told in the video to identify authentic chips. So I want to know if I can use these chips on a 12-0-12 Volts transformer which has a peak output of 34 to 36 Volts at no load and may drop to 33 Volts on load ?
This is very weird, because a month or so ago, I tried a quad amp with 4 TDA2003 that resulted to be fake, because I wasn't able to make them work at all. After buying some more in my local store, the same circuit came to live. The weird part is that the REAL TDA2003 looked just like this fake one you show AND the fake one looked JUST like the real one in this example. So I think been able to spot a fake one is finda more difficult.
There's been times for me the fake part saved a machine from land fill. Either the OEM part was gone long ago or the machine's price point for repair was to high.
@@GeorgeGeorge-xj2bc No it is not "meaning" that. It simply means the part is not an original part made by the company they claim it is from. It could be complete shit and not working at all, or it could be an accurate reproduction that works perfectly.
In your experience how well do counterfeit components hold up to dimensional tolerances? In the past i have had occurrences where a batch of power transistors have been a fraction too large, and they then won't fit the pcb correctly. IIRC the return rate for that batch of products was quite high. All the items came from the same major UK supplier, but I'm sure that it is very easy to get counterfeit products into the genuine supply chain. After that one batch of components the size went back to "normal", but when I complained the supplier just put it down to tolerances.
I got stung by fake parts once myself. So I built myself a device for identifying them. A high-precision ultra-low-voltage power supply and meter. You can step the voltage across a diode up in 1mV increments from 0.001 to 5V and monitor current. Just the thing for telling the schottky from the not-ky. The particular sting for me was a regular rectifier diode which had been re-labeled as a schottky. Did not go well when I tried to use it in a buck converter.
This is way off topic, but I bought some 74HC04's that were made by SunMoon, that have a weird non-HC-like input response. Cannot decode the part of their datasheet on the issue. Chinese sellers often will not disclose the manufacturer of the parts they sell. Is there a web-site that lists the best/worst Ali and Ebay IC suppliers?
Thanks for the videos John, good stuff. Been building power supplies, guitar tube amps, now on the gain clone/chip amps. I need a spectrum analyzer capability that does not break the bank. I see that your Rigol Oscope has this function. What are your top few choices on how to achieve this functionality? Thanks, George.
I know with some transistors they get a wafer of a similar type to make a near copy. 2SD756 and 2N5551 for example. The give away is the gain is very different. John. I liked your LM386 with booster transistors. I very much liked how it failed to work and why. If this TDA2050 is a re-badge it would be interesting to know what. I agree that it is unlikely to be new production.
I have some, purchased on Ebay from China. Just to play with, luckily! I think now I will derate them, expect 1/2 the performance (and no short protection!! UGH!), for 'toy' use only. Fun to play with, but clearly if you want performance, you must seek out the real deal. Thanks for the vid!
+JohnAudioTech oh man..that's ridiculous... hope that ST might come up with another series of audio ICs..these tda series are not at all available.. and if available they are all counterfeits
@@sayantanmaiti2513 The LM1875 is available to replace the TDA2050. I think ST has a new market, stereo/quad amps with 12V-supply for automotive applications. You might be interested in the TDA737, a 2x30W into 4ohm class AB amp that Mouser is selling for £2.86 at 1-off pricing.
@@RexxSchneider actually both 2030 and 1875 both have their counterfeits filled in the market. Originals ones are there but very costly. And if you talk about right price and original, then yes, the 73xx and 74xx series of TDA ICs are currently being manufactured
Dear John,do you maybe know what is first integrated chip amp.Who made it and so on.I have read it somewhere but cant find that article now.Love your videos.Keep up the good work.
The first audio power amp on a chip I can remember using was the LM380 back in the 1970s. I have an application note (AN-69) from National Semiconductor dated December 1972, but I can't be certain it was actually the first monolithic amp, although it must have been close.
If you rub the front firmly with a cotton bud dipped in isopropyl alcohol, the writing sometimes rubs off of a fake one. Also, if it's a used original, the legs may have been dipped in solder to disguise the fact that they have been desoldered from a board (legs can never be cleaned perfectly so they make them look uniform by coating whole leg)
I bought some LM675 power Op-Amps off ebay a while back… Bought some LED displays at the same time… OVER 50% failure rate... Needless to say, I never even bothered using the LM675s, and switched instead to an American surplus supplier who had some NOS power audio drivers that were made in Japan back in the 90s
Hi John,, Thank you for doing this video,,, It is a very big help,, Do you know if there is a way to tell the difference between Fake ( chinese ) J201 Jfets and the Authentic J201's? Thank you John.
It's quite difficult to make a counterfeit JFET that doesn't behave like a real JFET, so the main issues will be whether the Idss and Vgs(off) lie in the (wide) range of permissible values for the real part. You're normally going to have to test the part for at least one parameter (usually Vgs(off)) because of the difficulty of setting bias levels for a JFET amplifier or ensuring cutoff for switching applications. A cheap transistor tester will give you a reasonable idea. As the J201 is a 40V part, you may have check that it will stand the supply voltage if you're using more than about 25V, particularly the breakdown voltage between gate and source when the gate is very negative. For low voltage supplies, it's very unlikely to be an issue.
The LM1875 kit on one of your older vids looked iffie to me also but you showed it worked just as good.. Im getting one of those lm1875 kit for a project but powering it will be a little complex for me..
I decapped one of those 1875 ICs from the kit and the die was 10mm square which matches the real deal. Performance matches as well. Some sellers of those kits say the chips are used (probably pulled from boards of a canceled product).
Any chance that ST changed the process size at the end of the products life to get more yield from the wafers? i.e. is it the same die shrunk down or is it a completely different die? If it's lucrative for the Chinese to spin their own die and manufacture these that's just crazy. Then again I know they have done so with some other chips.
I don't know much on the business end of things, but I can't see them redesigning the die on a product near the end of its production. The output transistors take up a large amount of die space and can't really be shrunk. I think the Chinese are getting a "lesser" die from somewhere, putting it in a package to counterfeit.
@@JohnAudioTech Perhaps the Chinese are running these fake dies off junk wafers or such or along with other fake chips to make the process worth it. Crazy stuff.
I have used few fakes for a while, if you don't push them to the wall they work fine. At the time I bought those I didn't have access to the shop which sells genuine parts, although some that I've got look just like your genuine IC, maybe few of them are.
When I was looking up my chips tda2030; I found on the web that the fake ones have more plain Chinese looking fonts. Whereas, the authentic ones have a special font that is difficult to replicate by the counterfeiters.
This also valid for other versions: 2003/2020/2030 too, cuz i got 2 TDA 2003 just to try out for fun and hear how it sounds "vintage" amplifier ic and both blew up after a few second, and i only applied 12volt and 4ohm load. Also it sounded like a chinese amplifier. There was no defect in the circuit whatsoever, everything was double checked before applying voltage to vcc pin My dad had a defective TDA2003, caps was all bad too, it sounded very distorted, and the ic ran for a decade, it was a yugo car radio, but that ic also had wider legs near the ic, also had that blob in the black spot, unlike the counterfeit ones i got for 2euro each.
I assume vendors like Mouser, Digikey, and element 14 would be safe to purchase from, does anyone know of any others that sell authentic parts! I was about to make a purchase from "odd wires", not sure now
Presumably only buy from companies big enough to deal direct with manufacturers. I can't believe the hobbyist market is big enough to produce chips for - might be big enough to fake in the sense of adding false markings to some other part. That raises the question of where these parts go - equipment - how do you guard against equipment with fake parts in. In this case probably a big market for low cost / poor performance audio equipment.
I've used them in repairing early 70's Pioneer stereos, they seem to work well. I believe Sanyo stopped making them in 2003 so originals are hard to find and I just automatically assume none for sale today are originals. I usually get two module sets one set matching the original module and the other matching the specifications but at the highest power rating in the same family (check the datasheets). Question I've always had was just to what degree are they knockoffs? If they bought the equipment from Sanyo back in 2003 and are using that same process are they really knockoffs?
Also not to mention that If we are repairing customer's equipment and don't notice that some are fake chips/parts... Mmmm! Very costly for us and may potentially be dangerous in some cases for them. Scary thoughts all round.
Tda2003 work only single supply at 18vdc. Tda2030 work with and without Dual power supply. and out more power on split supply. on single supply at same voltage the power is same
Just got myself some - assumedly fake tda2050 .. also used a whole batch of tda2030 - good enough for learning. I certainly do buy real parts from Mouser thought. Especially anything like an LM3886
I wonder if the fake chip would have reset on a power cycle. There are some stray capacitances in the perfboard that could cause a lesser chip oscillate. Also the oscillation frequency is high enough for the zobel capacitor to pass enough signal to the zobel resistor and cause it to overdissipate. This was a very informative video that shows once again that the fake devices are more likely than not to be inferior. If the Chinese were to make counterfeits that were worth a damn, us Yankees (and everyone else) wouldn't be caterwauling about it. Instead, you show, yet again that these fakes are inferior, in this case delivering about -3dB compared to the real part. That's about half the power. I imagine that with a lighter load (8Ω or more) it might behave a little better, but still. What is really disturbing is vacuum tubes. The idea of a device with 400 volts across it that could fail short is unsettling at least.
i have build several amplifiers use tda 2003/2040/7294/1514a and many others using a couple of darlington like tip142/147 , all work very well but it's hard to calculate a good protection for the output transistors :( for exemple my DIY 2x200w using darlington if i short circuit the output...poor tip142/147 burn immediately...do you have any good idea for this? thks a lot ;)
@@madjidhamdini8114 If there's a chance of shorting the output and you have expensive output transistors, then you really have to have output current limiting. Monitor the current through small emitter resistors and use that to short out the drive to the output. It's a common enough circuit that you should be able to find numerous examples to copy. To protect speakers, it's a good idea to have some sort of fast-blow fuse in-line with the speakers anyway.
I'm listening to this on a powered speaker I made using a TDA2030a kit I got on ebay from China. I'm guessing they are fake. I did pick up a few more from a US seller. I'll have to compare them to the Chinese one.
Before I knew better,I ordered about a hundred bucks of Toshiba audio outputs from China. My amplifier kept self destructing. I separated a Toshiba from China and an original, and I could not believe the surface area difference n the two chips. I know better know.
thank you sir-actually my speakers are cheap- but still is there any easy method protection for speaker ?--sir-if i use -25-0-25 v max voltage on fake and authentic chips-power output would be how much ?--is there overtemperature protection in fake chips-sir i like your videos very much-i am enlightened in electronics world by your videos-sir i have lot of doubts regarding amp,preamp etc -can i get your email so that i can send you -if possible
Hi. I buy and use the fakes. Partly because they're cheap and available in well-designed modules. I'm not interested in short-circuit protection. And, of course, the difference between a 20W amplifier and a 13W amplifier is... almost nothing. If I drive a 95dB efficient speaker with a 5W amp it fills a hall. I have an LM386 0.2W amp that is louder than usable in my living room. I just don't see the downside of these terrific, cheap fake tda2050s and the modules they come in. $3 for a 13 W amp on a bord with screw connectors, power supply and heatsink (good for about 3W). What do you want for $3? A miracle?
I remember an electronics magazine here in the UK warning of counterfeit semi -conductors being sold from china like the 2N3055 for example,using a less powerful transistor re-branded,and how when stocks were low from the actual manufacturers some companies and i mean reputable distrobution electronics suppliers,were buying in their naivety from chinese distributors and then to western consumers etc under rated re-branded components,which were not up to the job intended. Sneaky little bar-stewards,mind you not all chinese sellers sell fake stuff i have to add.....but!!!.
Bought a mono kit with a Lm1875, scary it looks exactly like the left one, same front and back side and it got scratch marks on it. As if a file took the sharp edges in the metal away. Not even copper But it says lm1875 on it, must be fake then....................... F****
The LM1875 has a different package style than the TDA ones. These kits claim the ICs are recovered parts and may have scratches on the package. If you file a corner of the metal tab, it should be copper. You still might have an authentic chip.
@@JohnAudioTech thx John, But on your video of the Lm1875 the sides are straight and does not have that half circular cut on the sides. behaps I should take the ultimate test.....short the output ;)
How do they get their hands on the original die mask, in order to scale it down? Do more videos on how to spot fakes and their internals. It's a fascinating subject.
It isn't scaled down. They used a die from another IC of lower power. If you look carefully you can see the layout is different. What would be interesting to me is to find what IC that counterfeit chip came from.
if you google it, it's the 2nd link. www.shine7.com/audio/12ax7_pre.htm there's also another site that has another implementation. diyaudioprojects.com/Tubes/12AX7_Preamp/
And the things are getting worse in the field of television repairing and even worse in computers from the tons of fake or used with malfuctions bga chips and looks totally impossible to find a new unused genuine bga chip to successfully complete a repairing.
I don't believe they are developing anything. They buy smaller, less powerful chips from a supplier and then package and label them with numbers from a more power chip to create the counterfeit part.
@@JohnAudioTech And even then, they are quite possibly getting those less powerful chips as "seconds" - i.e. ones that don't even meet the spec for the lesser chip.
most of pre build tda20x0 amp kit using fake power ic. fake ic produce lot of static noise regardness of input. do not try to debug your circiut just pull out fake ic and put genuine one.
The manufacturing process for genuine ICs results in a proportion of below-spec chips being produced. Those "seconds" can either be thrown away, or somebody may cheaply purchase or otherwise obtain them by some means. Either way, the fabricator probably doesn't care. Those below-spec chips can be sold as good, or re-labelled as a higher spec part and sold for even more profit. Even if the profit is small, it adds up if you sell a lot of them.