Wow man, you've really put the work into this one! The awesome part is you are teaching us along the way. Looks like this little amp is going to be a beast!
Love both ypur channels to bad you couldn't combine both channels to make something really cool. With his know how and extensive knowledge on amps and your knowledge and know how on testing could be one badass ( excuse my language ) channel. 100% gold idea right there.
@crewchief No that's not what they are doing here each have their own channel,one test power the ither a mad scientist. im talking about having both on one channel. But i know you felt like you know what im talking about and thank you for the incorrect reply
Holy shit Sam! No wonder it's taking a minute! That is wildly complicated! I'll bet it's 6 or 7k+ before it gets too hot. The heat sink is gon get some work. I'm stoked man thanks for the update
So, you are able to figure out these issues and the entire R&D team can't. This tells me that you definitely know your shiznizzle. I absolutely enjoy your channel and can't wait for the unveiling!!!! Many complex! Such thinks! Wow.
I wouldn't worry about the litz wire bro, even 30k amps don't use it, I can't think of a single amp that does. That core is a quality one and will do the heavy lifting, you can tell because of the lack of wire it needed on the 8k. Love your work!
Yo Sam! This is the best series I've seen and the one I've been waiting for you to do! I always wondered why you didn't make slight changes to some amplifiers to improve reliability or power since you know so much about how it all works. I didn't know it would be so complicated and so difficult to get it all right. Im really looking forward to your part 3! I would really really love it if you worked with a Korean build house to design and have built your own line of original design amplifiers. You should consider doing it man. Start your own brand that is reliable and efficient, and obviously puts out gobs of clean power. I'd buy one in a heartbeat, and you have hundreds of thousands of potential customers once Derek puts your line of amplifiers on the Dyno. It's a golden opportunity. Don't pass it up. It could change your life.
O and I’m waiting for the day for DD Audio or Kicker to hire you bro! I personally think Kicker would be awesome but it has been cool to see your progression over the years.
HEY MAN I AM SUPPER EXITTED ABOUT THIS SERIES!!!!! Off to watch part one> I really enjoy watching repair. But mod for awesomeness. What a great idea. Love your work. Keep on keepin on Brother LOL
I've been following this for a good while now. I would love to see this thing in action when it's finished. I was wondering if you will finish this wild and interesting project. Love what you do.
Very well done, so we were correct about the duty varying being the feedback sensing. I will be sending my audio pipe to Derick soon as well, we'll see how much my modded Audio Pipe APMI1500 .1 does now.
So technical, I can't completely follow some stuff but what I can follow is that this project is freaking awesome!! I can't wait to see you hairtricking with that little thing. Enjoying a lot Sam, thank you for the content!
I'm really into audio electronics that much but not really knowledgeable about these kind of stuff, I always thought I could change the components on my amps to make more power and just by watching this, I feel like I had an electronics degree 🤣 cant wait to see to the final stage.
Awesome in all videos , unbelievable freak dyno amp test I think a tons of watt coming out with bigger fet can handle more amps. Great job about reverse engineering 👍
Hi Sam, nice job indeed; some comments about your mods: 1. 200v rail is the absolute maximum limit for irs2092. It may run pretty fine at idle but when you ask for some serious current trough the load overvoltage spikes will appear on your fets. The spikes WILL kill your irs2092 at 200V. Stay below 170...180v to be on the safe side. This can also explain why your irs2092 is killed sometimes at power on. 2. Dead time: from the waveforms you measured the fets are hard switching at idle and means the dead time is too short. You said it is set to minimum (DT pin) which is a very bad choice. Since I don't think you are too concerned with the Thd of this amp simply put it at maximum by changing the R around DT pin. This will remove some stress from the output fet that are clearly slower than the original ones. 3. Output inductor: good idea to increase the value but don't use the transformer core for it, it will never work. The transformer uses an high permeability core (2000..5000u) while the output inductor uses a low one (10..100u). Winding the output chocke on the transformer core will make it instantly saturate and you fets will be broken right away. Increase the turns on the original core or stack 2 identical cores if you need more inductance. A suitable core will be something like T130-2 or T157-2 or other distributed gap cores like Kool mu, sendust or MPP types. 4. Duty cycle change with rail voltage: this is something not intended by irs2092 and I think is a kind of protection to make it clip earlier at high rail voltage like you clearly explain. Check whether there is a resistor going from the positive rail to the feedback circuit around pin 2 or 4 of irs2092. If you find something like this simply remove it. The output idling voltage shall always be vrail/2 whatever vrail is. If this is noz the case means there is something addind DC offset to the feedback and has to be removed. 5. Litz wire: you don't need it at 25kHz and I don't think you can push the PWM frequency up too much with these beefy power supply fets you choose. Use many small wires in parallel to make it easier to wind Great job so far... I hope you the best when you will start asking some serious power from it.... Good luck
Thank you sir! Some responses to your tips: 1) Yes! I noticed this a few days ago when studying the datasheet. Bugger. This is a shame as I could have used some 200V output fets with higher power handling... and it does limit how much power I can ask from this amp. A shame, but oh well. At least I know the reason the chips were dying! 2) I'm sorry I may have explained this backwards. The chip is currently set to MAX deadtime by default on the board. I cannot ask it for any more deadtime. It would be nice if I could... but DT pin is tied directly to Ground, if you check the datasheet for 2092 this puts it into maximum deadtime bracket threshold 4 (
@@barevids 1. 200v is the ultimate limit you have to live with it.... 2. DT is already at max... Ok it is not enough because you lose zvs at no load. It is not a matter of switching speed is a matter of increased output capacitance of the fets. Increasing rail voltage will be even worse. Two ways to solve it: use mosfets with lower Coss (you don't need 300v btw.) or REDUCE the value of the output inductor. This will add triangle current in the fets at idle but the final condition is that the output capacitance of the fets must be discharged by the inductor triangle current during the dead time PRIOR fet turn on. Unwind some turns from your inductor and check the waveforms until you achieve this. This must be done at the highest possible rail voltage. Failing to do this will result is blown fets under power. 4. Sure but I don't know where to find the schematic of this amp... If it available somewhere I can have a look at it
I was looking at amplifier boards that you can buy that doesn't have a power supply part of the board. You can build a full bridge rectifier with a filter capacitor hook it up to a 120 VAC wall socket and get it to output around 170 VDC going to the rails it would be cool building your own home audio amplifier with high output. You also could use a toroidal step up transformer and have more voltage if 170 VDC isn't enough.
This is exactly what I want to do to my MTX7801 and my brand new Orion XTR 2500Dz, I have the equipment and electronic experience to do so BUT I don't have the experience of what other components (that you pointed out, and what waveforms to look for). This modification is really exciting to watch what problems that you come across and the remedy that you come up with. As Wiledy Coyote said "GENIUS SUPER GENIUS!)
Awsome video . Love the explanation and the information. Thank you for taking the time to go through,start to finish. Its very interesting to see this amp upgraded internally. Can't wait to see how much power you manage to squeeze out of this little amp
Very curious to see how the traces handle. Always makes me laugh to see a ~200 figure on those tiny transistor legs but it is a very short run. Either way, all the solder and everything else increases resistance and learning how that stacks up would be cool. I wish you were closer, we have a few high res (lol high res IR is .5-1MP) thermal cameras you could use. Maybe someone local who works with drones or industrial apps has one.
amazing! dude your on track for being an amp designer. there is a guy GR Research on youtube, he upgrades the Xovers of premium 2 channel loudspeakers. you could be the car audio version of him
Dude. Your a freaking genius! Why aren't you working for these companies designing boards?! Soooo looking forward to part 3. I'm a big fan of A.P. My whole audio build is A.P. the D is an APCLE30001.D and the 4 channel is an APCLE2004. But honestly I was disappointed when I saw them copying the Brazilian amp.
@David Long I was a CNC programmer and Machinist for years I stopped recently and now build sub boxes in my basement. I've got a Taramps HD 3k that this Audiopipe is a copy of. I'd take the Taramps in and use it to design the heat sink with the Barevids "B" with the speakers in it right on the top, I'd also add more thickness to it and machine it from a solid block of aluminium. After all this work, its not much of an audiopipe its a Barevids amp
Talking about going down the rabbit hole. I love your videos your enthusiasm keeps me hyped. Yes big D I love his stuff. I’ve made amplifiers so I understand what you’re doing. I just have to wonder if anyone else does? Keep up the cool work I definitely will keep watching. I never got an answer back do you know what the comb is Nguyen talking about Speakers?
How is is going with the Audiopipe brazilian clone amp you've been tuning up for testing in the van and sending to Derek for the dyno test? Please let us know if and when you get to publishing part 3. We're looking out for that. Thanks.
Greetings, sir: I have question, in hopes that you may be able to help me out. four years ago; I bought an amplifier at walmart. It was the "Dual XPR84D, class D 4-channel amplifier. It claims that the total power out put is 1,000 watts, however, it uses two 25 Amp fuses (50 Amps combined) at 12 - 15 volts D.C.. Don't get me wrong; I'm actually pretty impressed by its perfomance even today (May 12, 2022), but I was just curious if you were able to do a dyno-test on the aforementioned model that I have. Thank you for your time and consideration. P.S. I enjoy your videos and appreciate the quality of your videos, thus, I trust your observations on the products that you put to the test.
I know just enough to be scared to even attempt a minor repair and be confident that im not gonna start a fire....because I'd most likely burn shit down or totally destroy something that could've been made functional by someone not me.. wish i had your knowledge and skill in this area of electronics
This is an awesome video! it's even cooler when you learn how amplifiers work in class just to see that if you change the resistance in the voltage divider portion of the cct actually affects the output impedance to the stage. and just curious what channel will the other video be on?
