LM1875 as substitute for TDA2050

[suzyj]

Hi guys,

Being a newbie guitar player, I was fooling around with my amp (a Fender Frontman 15G) and blew the bejesus out of it. It's my own fault, the poor thing was never designed to have a line level component (my CD player) plugged directly into its aux input. It was impressively loud though

In any case, I killed the output amp (a TDA2050). This in turn caused the power transformer to overheat. Luckily, the fuse escaped unharmed.

I'm yet to find a replacement TDA2050, so I figured I'd bung an LM1875 in instead, and replace the power transformer with a spare twin 15V secondary 20VA one I had kicking around.

I figured the LM1875 was as close to a pin-for-pin replacement as I was likely to find. Certainly it operates from higher power supplies, and is good for similar power to the TDA2050.

However now when I power it up, the amplifier is somewhat less than stable. I can set up audio frequency oscillations, and get really interesting sustain on some notes.

I'm thinking the best place to start is to replace the output shunt RC network with something a bit more substantial (it's 5.6 ohm in series with 100n now), and see where I go from there. There's also a pair of back-to-back diodes with a 100K series resistor providing positive feedback, which has me scratching my head.

It's entirely conceivable that the interesting effects are coming from further back in the circuit, as well.

Anyway, does anyone have any thoughts on the validity of the LM1875 as a replacement for a TDA2050? Any other leads to follow?

For reference, the schematic of the amp (though not showing the auxiliary input) can be found at http://www.fender.com/support/amp_sc...ematic_68F.pdf

Cheers,

Suzy

[consort_ee_um]

In general the LM1875 can be used to substitute the TDA2050. You have to change the RC network on the output though, and you will not get the power into 4ohms of the 2050.
The Fender circuit is so unusual that I would say it is unlikely to work with the 1875 without tweaking. You could try removing the 100k positive feedback resistor to see if it is stable and then put say a 470k back to get the desired effects.

[suzyj]

Hi guys,

Sorted. I followed the following steps:

1. add 1 Ohm/0.22uF shunt to output. Slightly better, but still significant oscillations.

2. Remove CR5, CR6, (diodes from output to supplies) and R39 (positive feedback resistor). Now significant (10V) output DC offset.

3. Short C23 to provide DC feedback. DC offset gone, but still breaking into oscillations occasionally.

4. Bypass VCC and VEE to ground at power amp. There's no bypassing near the chip, and it's a few inches away from the main power supply caps (with relatively narrow traces). I used 10uF 40V electros with 100n poly caps across them, as that's what I had to hand. This removed all the oscillations, and the output is now nice and clean.

Steps 1 through 3 were probably unnecessary, but I left the mods as they were.

Cheers,

Suzy

[neutron7]

Originally posted by suzyj


In any case, I killed the output amp (a TDA2050). This in turn caused the power transformer to overheat. Luckily, the fuse escaped unharmed.

LOL!!!! about the fuse

[teemuk]

Originally posted by suzyj
the poor thing was never designed to have a line level component (my CD player) plugged directly into its aux input.

Makes me wonder but isn't this why aux in is there for...? Are you sure you located the problem to the right source? In my experience it's quite hard to destroy TDA chips with too much output power since they have a built-in thermal protection. Usually they start to attenuate notably before any serious damage happens. At this point the user should know to reduce the volume by him/herself. Did you do anything else to the amplifier besides plugging a CD player to the aux in or did Fender just equip it with a super wimpy heatsink?

Anyway, nice to hear you already get sound out of it with a different chip.

[suzyj]

Originally posted by teemuk


Makes me wonder but isn't this why aux in is there for...? Are you sure you located the problem to the right source? In my experience it's quite hard to destroy TDA chips with too much output power since they have a built-in thermal protection. Usually they start to attenuate notably before any serious damage happens. At this point the user should know to reduce the volume by him/herself. Did you do anything else to the amplifier besides plugging a CD player to the aux in or did Fender just equip it with a super wimpy heatsink?

Anyway, nice to hear you already get sound out of it with a different chip.

I'm reasonably sure about the cause of the problem. The heatsink on the TDA2050 was only a 2" by 3" aluminium bracket. I've bolted a reasonably substantial heatsink to it now though. The transformer was on the small size too. I'd doubt it was bigger than 10VA.

Cheers,

Suzy

[Gold_xyz]

Hi,

I have the same problem but i need replace a LM1875T
(On My Marshall MG15CDR) with a TDA2050 because LM_xyz
is hard to find.
I have not burnt IC but I have only broken him during
the numerous "tuning" on PCB.
It seems me that LM1875 are now available with a different package
but on national-datasheet they are not reported...

Bye

[Jpoole908]

I think I may have the same problem with mine. what is a TDA2050, what does it look like, and where can I get a replacement?

[cpemma]

Originally posted by Jpoole908
... what is a TDA2050, what does it look like, and where can I get a replacement?

Looks like a TO-220 transistor with 5 leads, comes in vertical-mount and horizontal-mount leads.

It's ridiculously easy to find a datasheet, type 7 characters into Google.