TPA3118 60W Power Amp Module

[Paul.Ruby]

Have been comparing several versions of active EQ (Tone Tweq, Cold Turkey, Box 'n' all, 6-band) and keep going back to the volume-only preamp. There is something about the active EQ I don't like. It's a bit unnatural feeling in comparison. But I want a bit of tone control. So, I built a tried-and-true passive tone stack with Marshall values. This really feels right. Sounds right. It's a guitar amp. This includes an option for a bright switch but I'm going to build without the switch and use a 120pf bright cap (C4), leaving out the switch and C6. Spent the morning getting the tone stack and bright cap chosen just right.

The expressPCB layout has two versions. I don't seem to be allowed to attach a zip anymore... ?? One with pots on copper side and one with them on component side. Building at home on single sided board makes the copper-side version very difficult to solder. I learned this the hard way, but I have one instance of that built up and in a case. Pictured here has the pots on the component side. This is easier to build but careful not to use components that are too tall. Note the 1uf cap laying on it's side and use of a surface mount electrolytic cap with it's base removed to make it through-hole. This will fit in a 125B case. I've ordered some right-angle header pins since can't use regular headers with this board. I might just hard wire this one. The LED resistor is not installed because every LED I try needs different current to get right. I still have LEDs from the 80s (I'm old) that draw 20 times more than this whole preamp... and modern super bright that seem to turn on if I get them near an outlet... Still too bright even with a 1M resistor.

 

[stevel]

I've been burned by those fake 100 watt boards too. Both from Amazon and Aliexpress. (I've been able to get refunds though, so I guess the sellers know they are scamming us). I had previously gotten a few that were real, and they look the same on face value but they certainly aren't. But those older boards are great. I ordered one of the boards with the screwed on heatsink and will report back when it arrives.

I've been using these little amps (Blue 3118 boards and the previously purchased 3116 boards) connected to a 1x12 guitar cabinet as my main rehearsal amps for a while now. I drive them with either an HX Stomp, an Iridium, or recently, a NU-X Amp Academy. For gigs (there haven't been that many), I use a more powerful TDA7498e based board or even a pre-built amplifier based on those chips.

I wish these sellers would provide proper specs (wattage @ impedance @ THD) instead of just quoting watts, but regardless, these TDA7498e boards are pretty loud. I built one from a baord, but I also have a few pre-built stereo HIFI amps based on these chips. I just use one channel, and they need conversioon cables (banana plugs on output and RCA or 1/8" on input), but for the money, they are hard to beat.

ex: Amazon link for 7498e amplifier

 

[Willybomb]

Well, we just played a gig on the weekend using the "100 watt" 3116 version as a bass amp to power a 4x10 cab using from a Helix. These things are plenty loud (as we've all been saying) but seem to work best with a pedal/preamp feeding a buffer before the amp.

 

[Paul.Ruby]

Yes, these amps take line level. All of the methods I've posted above work well. And certainly need a good preamp as well. I use an AMT M2 as my preamp then the "volume only" version above on my pedal board. That's the core of the board. Been playing gigs continuously with these.

 

[Mike McLane]

Anybody have misgivings about using these boards for serious gigs? Was it earlier on this thread or another where the TPA3118 (max 24vdc) was sporting 25V caps that burned up. Can you really get a hold of a board that has some "umph" to it? Also, Chuck Bones had a fairly thoughtful piece about the fact that all power supplies are not created equal. That something for this situation (like the TPA board itself) needs to be carefully chosen. Just because your laptop psu puts out 19vdc @ 3+amps doesn't necessarily make it ideal for this. I've been looking at the ICEPower boards which, while much more expensive are also MUCH stouter. . . . with power supplies, buffers, etc. already designed for and integrated on the boards. Look to be pretty close to plug and play. Same units as used in the Seymour Duncan PowerStage and Fender Tonemasters.

 

[Paul.Ruby]

The other guitar player in my band and I have been doing it regularly. I drop a 24V supply down to 23V with a series diode (it's posted in this thread) and I used an oscilloscope to show my supply has no overshoot and is well behaved. No problems... Also, I've been using the 24V output of my cioks DC7, which is awesome.

 

[Paul.Ruby]

I should add... Since these are easy to build and small, we have spares... Just in case. And, I made the guts modular so it's easy to swap a board if needed. That's all shown in my posts above...

Update after more than a year of gigging these, I've not had any blow up.

 

[Robert]

Not sure if it has been mentioned previously in the thread, but this isn't much different from how the Friedman BE-Mini works.

 

[Mike McLane]

I'm not an electronics guy. I just want a 125B PCB that I can assemble then plug in and go. All of this stuff looks and sounds great, but I don't have the technical chops for it. Mr. PedalPCB initially teased us with this. . . . .



That's what I'm lookin' for.

 

[fydo]

My TPA3118 arrived today and I am excited to see what is possible with it. Watching this thread, for sure!

 

[Robert]

I couldn't get the correct 1/4" jacks for a while (they have to be Neutrik, the clones wont work) so I put this back on the backburner....

Maybe they're available now, I'll have to go check. That's really our only hold up.

 

[Paul.Ruby]

I'm not an electronics guy. I just want a 125B PCB that I can assemble then plug in and go. All of this stuff looks and sounds great, but I don't have the technical chops for it. Mr. PedalPCB initially teased us with this. . . . .

View attachment 28723

That's what I'm lookin' for.

Want to buy one of the ones I made? I have... Many... Because I was comparing different preamps.

 

[Elijah-Baley]

Meanwhile, I received my first TPA3118.
I guess I have to wait the end of the summer to test it: power up with a 1-spot 1700mA PSU and then another cheap PSU for pedal, a RockPower 18v 2000mA. I'll use a 15w 8" speaker.

Are there any risk of damage something with those stuff?

 

[temol]

Before connecting the module to the speaker I'd user 100W 8R resistor as a load and measure output power (with oscilloscope).
15W RMS equals around 15.5V p/p (into 8R load). But it's pure sinewave. If you clip the signal (in preamp or by overloading input of the amp) then clipped 15V p/p is no longer 15W RMS. 15V squarewave = 28W RMS. 60W RMS equals 31V p/p (8R load).

 

[Mike McLane]

Roberto - Will the PCB you're doing accommodate the 3116 or other higher powered modules? I play bass fairly often and would love something that will do for a small club.

 

[Robert]

Roberto - Will the PCB you're doing accommodate the 3116 or other higher powered modules? I play bass fairly often and would love something that will do for a small club.

I'm not sure, I haven't looked at the 3116. My initial guess is probably not as a direct plug in replacement.

 

[Mike McLane]

First things first. If you can get the 3118 flying I'm good. Having 100W @ 8 ohms would be probably be great for a small bass rig, but the 3116 @ 50W @ 8 (?) ohms would probably suffice as well.

 

[Elijah-Baley]

Meanwhile, I received my first TPA3118.
I guess I have to wait the end of the summer to test it: power up with a 1-spot 1700mA PSU and then another cheap PSU for pedal, a RockPower 18v 2000mA. I'll use a 15w 8" speaker.

Are there any risk of damage something with those stuff?

I forgot, but the speaker is 8 Ohm.

Before connecting the module to the speaker I'd user 100W 8R resistor as a load and measure output power (with oscilloscope).
15W RMS equals around 15.5V p/p (into 8R load). But it's pure sinewave. If you clip the signal (in preamp or by overloading input of the amp) then clipped 15V p/p is no longer 15W RMS. 15V squarewave = 28W RMS. 60W RMS equals 31V p/p (8R load).

Thanks, but I'm lost, a bit.
Anyway, I didn't have any 100W 8R resistor, I don't have the oscilloscope.
I thought to test the power amp with the guitar directly into it, and the speaker at the output.

I checked again the datasheet, maybe I asked it before, but as I remember at 9v my speaker should be safe. At 18v I should avoid higher volume, because the wattage at 18 is close to 20w.
This is about the speaker.
What about the mA of my two PSU? The 9v with 1700mA and the 18v with 2000mA?
If I'm not mistaken, the mA consumption should be relatively low, right?

 

[Robert]

First things first.

Looks like the jacks are back in stock at Mouser. I'll get the ball rolling on this again.

 

[temol]

Anyway, I didn't have any 100W 8R resistor, I don't have the oscilloscope.
I thought to test the power amp with the guitar directly into it, and the speaker at the output.

There's a safe way, and your way. You have to choose
You can keep the volume down but one day someone can crank the knob to the max and then what?

Soundcard oscilloscope software https://www.sillanumsoft.org/prod01.htm
google "soundcard oscilloscope probe"

Btw - the measured value is relative. To calibrate the probe you need a reference signal, with known level. Also - if you have a probe with two diodes at the input - they would clip if the signal exceeds diode fv (0.6-0.7V for the 1N4147). So if the input signal is greater than 0.7V you have to adjust input potentiometer/trimmer (voltage divider acting like VOL pot) to bring the signal below the clipping threshold. Each change of the "sensitivity" requires re-calibration.