TPA3118 60W Power Amp Module

Heads up - the little blue 60w ones will die with 24v. One bloke found out on the DIYstompboxes(unofficial) FB group, and I had it happen to me when I tried 24v rather than 19v. The caps are rated to 25v, but clearly not really...
 
you will be OK with the 19v supply on any of those speakers as long as you do not try to play the lower watt ones too loud.
 
Ok, so it's enough keep the volume under 3/4 if I'll use a 19v PSU.

Maybe I could try to build a 12v regulator circuit, just to to build it and see how the TPA3118 will sound with the smaller speakers.

I have to wait to make some money to order the amp and the PSU.
 
I finished inserting the Simple JFET buffer btw the input and VOLUME knob and it seems to work EXCEPT the output has shift decidedly to be more "trebley" . . . less bottom end, sharper highs. . . giving it a somewhat thin sound. The shift isn't huge, but definitely less satisfying than without the buffer. I'm using an HX Stomp set for line level output so in theory the buffer should be unnecessary, but other front ends may require it so I thought I'd experiment and it's presence shouldn't hamper the HX signal other than its obviously is "EQing" the signal. Anybody have any thoughts?
 
I used the values prescribed on the Simple JFET Buffer build docs. I powered with a 19vdc computer PSU and ran the buffer power thru the 9vdc voltage regulator and cap values WillyBomb suggested above. The layout is attached.
 

Attachments

  • img20220119_16152364 (2).jpg
    img20220119_16152364 (2).jpg
    505.7 KB · Views: 57
I used the values prescribed on the Simple JFET Buffer build docs. I powered with a 19vdc computer PSU and ran the buffer power thru the 9vdc voltage regulator and cap values WillyBomb suggested above. The layout is attached.
I'm not using that 220n on the volume pot. Ditch it, and stick another buffer after the pot.
 
Ok, further updates:

I boxed up the big red 150w PCB in the usual manner - input, buffer, volume pot, buffer, amp - and found it way too noisy. The 24v supplies I've gotten have a high pitch, low volume noise in there with any of my amps so I'm using the 19.5v ATM, but it was pretty unusable even with that. I've replaced it with the 100w medium blue pcb. That deals with 24v fine, but the PS noise isn't acceptable. The little 60w boards seem to be the best compromise.

You might remember 100Lolwhaats? from earlier in this thread. I sent that to a friend in Melbourne to try out with his band.

He finally tried it out, and this is our convo:

- Did lolwhaats hold up?
- Did it ever! Sounded great!
- Sweet. What did you run it into? Keep up with the drums et al ok?
- Had the 1×12. I put it up on a bench to give us more floor room and it was plenty loud enough. Had it dimed with helix volume set at about 11 o'clock. Sounded awesome with the all the clean and dirty stuff. The only problem was the acoustic guitar didn't sound great. Must need a full range speaker...
- Well, that's handy to know. If the acoustic was going through the helix that would be a problem maybe. I don't think these things work that well with a guitar going straight in, at least that's my experience.
- Wasn't an actual acoustic. Just the variax into a patch with a Comp to boost the volume and a graphic eq. And some reverb.

So that might be something to look into down the track.




XH-M544-DC-12V-24V-150W-TPA3116DA-TPA3116-D2-Mono-Channel-Digital-Power-Audio-Amplifier-Amp.jpg_Q90.jpg_.webp
 
Schematic and PCB layout removed from this post in favor of new ones below...

I just built a few pedal-board amps in a 1590B case. Here's my preamp design, which runs off the same 24V power supply. Spent a lot of time futzing with C5 and C6 to get the top end just right without needing any extra tone controls. This shows a bright switch but I have it hard-wired on in this build. R1 sets the input gain and I dialed this in to work well with pedal levels. Goal was to be as simple as possible. Just a volume knob and maybe a bright switch.
I have a PCB layout for it as well... Attached zip is for expressPCB. Pic of guts below. The black coating is for insulation, not to hide anything... There's just a volume pot and I have the "bright switch" jumpered on. The volume pot is where a footswitch would usually be. You can just see the edge of it on the right side of this pic behind the preamp board. R5 is mounted on the input jack. R13 is mounted on the LED. C8 does not exist in my PCB layout but included in the schematic in case someone can lay it out better then I did.

VolumeGuts.PNG
 
Last edited:
I just built a few pedal-board amps in a 1590B case. Here's my preamp design, which runs off the same 24V power supply. Spent a lot of time futzing with C5 and C6 to get the top end just right without needing any extra tone controls. This shows a bright switch but I have it hard-wired on in this build. R1 sets the input gain and I dialed this in to work well with pedal levels. Goal was to be as simple as possible. Just a volume knob and maybe a bright switch.
I have a PCB layout for it as well... Attached zip is for expressPCB. Pic of guts below. The black coating is for insulation, not to hide anything... There's just a volume pot and I have the "bright switch" jumpered on. The volume pot is where a footswitch would usually be. You can just see the edge of it on the right side of this pic behind the preamp board. R5 is mounted on the input jack. R13 is mounted on the LED. C8 does not exist in my PCB layout but included in the schematic in case someone can lay it out better then I did.

View attachment 22915
Tops side pic as well... top.PNG
 
Last edited:
Why the output jack is made of plastic? It's have to do something with the ground?
Yeah, you can't ground the output to the enclosure. First google hit:

"That's because most amps and receivers these days have bridged (or differential) output. Both the negative terminal and positive terminal of bridged output amplifiers send a voltage signal to the speakers, so there is no common ground."
 
Last edited:
Why the output jack is made of plastic? It's have to do something with the ground?
Exactly. The speaker output lines must not be grounded. The negative of the input, however, is common with the negative of the power. So, I have the negative of the input to the preamp and the negative of the power supply directly connected to the case.
 
Since I can't leave well enough alone, I made a version of the PCB without the mounting holes since I don't use them. Both versions are in the attached zip file for expressPCB. I'm adding a 3A diode (1N5408) in series with the input supply line to drop the voltage from 24 down to ~22.5 since I see some reports here of the caps blowing on these little power amp boards. Updated schematic:
VolumeSchematic.PNG
Construction photos... Super glued the LED to get it down tight to the case and well insulated. R13 is soldered directly to the LED and leads are heat shrunk wrapped.
P1010674.JPG
Modular connections using JST-VH for power and speaker connectors and KF2510 for all the signal lines. Note this PCB is a bit earlier than the one in the attached zip file. The input and power connectors are too close together on this one. All resistors are 1/8W to squeeze this all down tight. The 10uf electrolytic is one of the short-fat types because the tall ones will touch the volume pot. Of course, all polyester film caps in the signal path with the exception of the two bright caps, C5 and C6 are ceramic. (This board has a film cap for C5.)
P1010675.JPG
Showing the speaker line run and how the amp board will flip upside down over the power jack. Power jack location is important as the lead end up between the JST-VH connectors and the 330uF caps. Tape is wrapped around the jack to avoid any contact with things on the amp board.
P1010676.JPG
All of the connectors in place...
P1010677.JPG
1N5408 added to drop input voltage a little and ensure correct polarity at the input.
P1010679.JPG
Heat shrink the 1N5408.
P1010680.JPG
Wire nut all the power lines together. Can solder and heat shrink, but this lets me take things apart until I'm sure it's final.
P1010681.JPG
Snug fit. Fully insulated the board and the case with liquid tape.
P1010682.JPG
 

Attachments

Last edited:
As others have pointed out, must use a good power supply. I use the 24V auxiliary output on a Cioks DC7 and it's dead quiet. Best pedal board power supply ever made. I just got a cheapo 19V from amazon and it whines. Now I need to see if there is a remedy for that...

After some testing, the problem is direct power supply noise. The TPA3118 module itself is rejecting this noise. Very quiet until I attach the preamp. Found no evidence of ground loop or RFI getting into the preamp. Purely power supply noise. Solution is simple. The preamp draws very little current, so adding a 470-ohm resistor in series with the supply line to the preamp solves the problem. The RC filter of 470 ohm and the on-board 10uf resolves the noise completely.

Post above has been updated with corrected schematic and PCB.

Still have a problem. It's quiet with the cheapo power supply and a guitar plugged in directly and sounds nice on it's own. Fendery clean with the choice of EQ in the preamp. But, married with my pedal board at large, some of the whine comes back if I use the cheapo power supply. Still dead quiet using the Cioks DC7, of course. My guess is differential ground noise between the board and the amp with the cheapo supply. Next step is to see if I can get a solid, clean ground across all the pedals with a cheapo power supply in the mix.

And again... Further testing. My choice of cheapo supply has a 3-prong, grounded power cord. The earth ground of the mains line is directly connected to the negative of it's 19V output. This allows the cheapo supply to directly insert it's ground noise into a ground loop when connected to the pedal board. The solution to this problem is to ensure one, and only one, connection of your pedal supplies back to mains earth ground. My pedal board is already earth grounded via the Cioks DC7. The added grounding via the cheapo supply causes the noise. Removing the earth ground on the cheapo supply mains line eliminates the noise. Finally clean and works really well.

Summary of my learning building these things...
  • TPA3118 is great. Flat frequency response (which is too dark on it's own, IMO). And rejects the noise of a cheapo power supply very well.
  • Need a preamp to boost guitar/pedal/instrument level up to line level to drive the TPA3118 well.
  • Fender-like EQ choice for the preamp works really well. Flat response from 20hz up to ~1kHz and then passive high-pass to boost the top end above ~1khz. (See C5/R6 and C6 in the schematic. Tweak to taste.)
  • Cheapo power supply can work. TPA3118 can handle it but the preamp cannot. Need a good RC filter or better regulator to get clean power to a preamp. With ~20V available, an RC filter is fine for a simple preamp like this drawing under a mA.
  • Don't allow more than one path from circuit ground back to mains earth from your pedal board supplies, especially if using a cheapo supply for the TPA3118. Do make sure you have ONE solid earth ground from your pedal board for safety, of course.
  • Cioks DC7 is awesome and can drive the TPA3118 using it's auxiliary 24V output. (I was already doing this with a quilter amp, which is also really nice).
Next up... I've ordered three different EQ boards from here to make some bigger, more full featured pedal board amps. I have a "Cold Turkey", "Box and All" and a "6 band EQ (Pot version)" boards that all arrived today. Nice timing. :) And I have a couple 100W TPA3116 boards to try out.
 
Last edited:
Just finished using a Cold Turkey EQ as the preamp. Have not put it in a case. Much more gain is needed to bring instrument level up to line level. Here's the schematic, which is just component changes on the PCB, all shown in red. Power supply is 24V, so choose capacitors to suite the higher voltage. coldTurkeySchematic.PNG
Summary:
  • Not using the foot switch, LED or R100 on the board. LED will all be off board, but you could change R100 to 47k and use the LED circuit on-board.
  • Main change is the input gain: R4=330k; R5=47k. This gets the signal up to line level.
  • Using TL072 but JRC4558 would work, too.
  • D100 is replaced with 220R to act as an RC filter for the power supply. I will have a 1N5408 off board for polarity protection and to drop voltage a little, just like the previous simpler board. This RC filtering is very important for use with cheapo 24V power supplies.
  • Replace C7 with a wire. I don't see the point of C7 but not important.
  • Leave out R14.
  • Change R15 to 10k.
  • Change volume pot to A100K.
  • Changed R1 and R3 to 1M. Not important but noise in a resistor is proportional to value, so I don't go higher than 1M for these bias things.
  • Left out C1. Not important.
  • I used film caps for the 1uf. With the removal of C7, there are no electrolytic in the signal path. All polyester.
Very easy build since the PCB itself needs no modification. Was actually much easier than making my own PCB for the simple volume preamp previously. The amp will fit where the footswitch would have been. Time case it up. Should easily fit in a 125B. Maybe a 1590B. Action Shot:
ColdTurkeyAmpGuts.PNG
That's the 100W TPA3116. That will not fit in a 125B...
 
Last edited:
The "Box and All EQ" also works really well with the TPA3118. Cold Turkey seems to boost midrange, which is cool and a bit more '80s marshall EQ vibe. Box is more fendery. Both work really well. Here's the altered schematic, again just making component changes to pair with the TPA3118.
boxAndAllSchematic.PNG
Summary:
  • Level POT changed to A100k. Did not need nearly as much gain boost to get to line level as the Cold Turkey did.
  • Use 25V or higher for C100.
  • Leave out the LED and R103. But, can include if you like and change R103 to 47k.
  • Change D100 to 220R for RC filtering of cheapo power supply. Lose about 2V, which doesn't matter at 24V.
  • Changed Vref resistors R100/101 to 100k to reduce current draw.
  • Jumpered C7 and C8 and left out R11. The DC bias voltage output from the op-amps is already at Vref.
    • This may allow a small DC current through the level pot due to non-zero offset voltage at the input of IC2.2.
    • So, you may choose to leave in C7 to be sure there is zero DC current in the level pot.
    • And 1uf film cap is fine for C7.
  • Changed C1 to 100nf. 1u is overkill.
  • Changed C2 and C9 to 1uf film caps because I prefer to leave electrolytic out of the signal path.
 
Last edited:
And finally, the changes to the 6-band EQ to pair with TPA3118. This needs a volume pot added. Primary gain increase is from R7 but also a bit from R3 and R9. Since the on-board LED come after D100 on this one, do not use the on-board LED since it is on the wrong side of the 220R resistor used to filter the cheapo power supply. This one is my least favorite of the bunch. Need to spend time tweaking the knobs to dial it in but it certainly has a crazy amount of EQ control. I like the original with just a volume control. Took me a bit to tweak the hard-wired EQ, but now it is plug'n'play. For more control, I like the cold turkey version as a #2. Box and All is also good but more stuff and bigger comes in just behind cold turkey.
6bandSchematic.PNG
 
By the way, as others have pointed out, these things make a big pop when powered on and off. A proper design would have a power on delay and detection of power off for instant disconnection. I can confirm that having the mute pins shorted eliminates all sound, including power-up/down pops. Power on delay can be short, maybe 100ms. Power cut detection must be fast.
 
Back
Top