Topics

TDA2822 ill treatment test!

 

Having used TDA2822 for decades and never had a failure of any kind even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got bad/defective/BY chips..

Cheers

--
Raj, vu2zap
Bengaluru, South India.

Jerry Gaffke
 

That's good news.

You've addressed the most obvious way in which it might fail,
though perhaps what causes it to fail is not quite that simple.

I wonder if hfsignals has stuffed the same part into all rigs thus far,
or has perhaps been switching between manufacturers,  

Jerry


On Thu, Mar 1, 2018 at 04:05 am, Raj vu2zap wrote:
Having used TDA2822 for decades and never had a failure of any kind even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got bad/defective/BY chips..

Joe Puma
 

Thanks for taking one for the team Raj. 👍🏼👍🏼👍🏼

On Mar 1, 2018, at 7:04 AM, Raj vu2zap <@Raj> wrote:


Having used TDA2822 for decades and never had a failure of any kind even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got bad/defective/BY chips..

Cheers

--
Raj, vu2zap
Bengaluru, South India.




w7hd.rh
 

I suspect the 2A limit on his power supply saved the chip!

Ron W7HD

On 03/01/2018 10:00 AM, Joe Puma wrote:
Thanks for taking one for the team Raj. 👍🏼👍🏼👍🏼
On Mar 1, 2018, at 7:04 AM, Raj vu2zap <@Raj> wrote:


Having used TDA2822 for decades and never had a failure of any kind even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got bad/defective/BY chips..

Cheers

--
Raj, vu2zap
Bengaluru, South India.




--
W7HD - NAQCC#7587 OMISS#9898 KX3#6966 LinuxUser#415320

Vince Vielhaber
 

The chip is only rated for 15 volts. Is it possible that the ones that are experiencing smoke are also getting more than 13.8v? Possibly a spike?

Vince.

On 03/01/2018 12:02 PM, w7hd.rh wrote:
I suspect the 2A limit on his power supply saved the chip!

Ron W7HD


On 03/01/2018 10:00 AM, Joe Puma wrote:
Thanks for taking one for the team Raj. 👍🏼👍🏼👍🏼
On Mar 1, 2018, at 7:04 AM, Raj vu2zap <@Raj>
wrote:


Having used TDA2822 for decades and never had a failure of any kind
even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip
blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this
volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got
bad/defective/BY chips..

Cheers

--
Raj, vu2zap
Bengaluru, South India.




Gary Shriver
 

I’m no engineer here but could that be to the 2A current of your power supply? Maybe the others are running a 20amp supply. Just a thought..
73-Gary

On Mar 1, 2018, at 4:04 AM, Raj vu2zap <@Raj> wrote:


Having used TDA2822 for decades and never had a failure of any kind even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got bad/defective/BY chips..

Cheers

--
Raj, vu2zap
Bengaluru, South India.




csmithuk0@...
 

Hi all,

Just a note on the TDA2822. It doesn't specify it has any kind of short circuit protection in it and looking at the schematic on the datasheet, it doesn't. On non-power opamps you tend to see small series resistors in the output chain which limit the total current the IC can deliver and stop it from exploding. This reduces power gain a little bit so on LM386 / TDA2822 they just don't exist. YMMV but if you short it, it'll probably blow up. If it doesn't, that might actually be the bad amplifier.

As for the cause of the failure, I've had a couple of (cheap no brand) coupling capacitors go short in the past. These drag the amplifier output down to ground which blows up the amp pretty quickly. Worth checking your capacitors before you stick another IC in there and watch that explode as well.

And another one: cheap 3.5mm sockets. Sometimes the contact positioning is terrible or gets mangled pretty quickly and the output can be shorted by just inserting a plug or turning it on. Again, boom! Check your 3.5mm socket for shorts.

I doubt that the IC would die if you overvolted it a little bit. There's usually a pretty big margin of error included. 

Incidentally I had an LM386 quite happily emitting a stream of smoke while working absolutely perfectly, well for a while. Not sure what happened there! Sometimes you get a dud.

On 1 March 2018 at 17:09, Vince Vielhaber <vev@...> wrote:
The chip is only rated for 15 volts.  Is it possible that the ones that are experiencing smoke are also getting more than 13.8v?   Possibly a spike?

Vince.

On 03/01/2018 12:02 PM, w7hd.rh wrote:
I suspect the 2A limit on his power supply saved the chip!

Ron W7HD


On 03/01/2018 10:00 AM, Joe Puma wrote:
Thanks for taking one for the team Raj.  👍🏼👍🏼👍🏼
On Mar 1, 2018, at 7:04 AM, Raj vu2zap <rajendrakumargg@...>
wrote:


Having used TDA2822 for decades and never had a failure of any kind
even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip
blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this
volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got
bad/defective/BY chips..

Cheers

--
Raj, vu2zap
Bengaluru, South India.









--
  Michigan VHF Corp.   http://www.nobucks.net/   http://www.CDupe.com/
                          http://www.metalworkingfun.com




M Garza <mgarza896@...>
 

Just curious if any of the rigs with blown chips had a regulator just for the board?  I am running a 19v laptop power supply and a 12v regulator for the board.

Marco - KG5PRT 

On Mar 1, 2018 11:33 AM, "Chris Smith" <csmithuk0@...> wrote:
Hi all,

Just a note on the TDA2822. It doesn't specify it has any kind of short circuit protection in it and looking at the schematic on the datasheet, it doesn't. On non-power opamps you tend to see small series resistors in the output chain which limit the total current the IC can deliver and stop it from exploding. This reduces power gain a little bit so on LM386 / TDA2822 they just don't exist. YMMV but if you short it, it'll probably blow up. If it doesn't, that might actually be the bad amplifier.

As for the cause of the failure, I've had a couple of (cheap no brand) coupling capacitors go short in the past. These drag the amplifier output down to ground which blows up the amp pretty quickly. Worth checking your capacitors before you stick another IC in there and watch that explode as well.

And another one: cheap 3.5mm sockets. Sometimes the contact positioning is terrible or gets mangled pretty quickly and the output can be shorted by just inserting a plug or turning it on. Again, boom! Check your 3.5mm socket for shorts.

I doubt that the IC would die if you overvolted it a little bit. There's usually a pretty big margin of error included. 

Incidentally I had an LM386 quite happily emitting a stream of smoke while working absolutely perfectly, well for a while. Not sure what happened there! Sometimes you get a dud.

--
73, M6XTG Chris


On 1 March 2018 at 17:09, Vince Vielhaber <vev@...> wrote:
The chip is only rated for 15 volts.  Is it possible that the ones that are experiencing smoke are also getting more than 13.8v?   Possibly a spike?

Vince.

On 03/01/2018 12:02 PM, w7hd.rh wrote:
I suspect the 2A limit on his power supply saved the chip!

Ron W7HD


On 03/01/2018 10:00 AM, Joe Puma wrote:
Thanks for taking one for the team Raj.  👍🏼👍🏼👍🏼
On Mar 1, 2018, at 7:04 AM, Raj vu2zap <rajendrakumargg@...>
wrote:


Having used TDA2822 for decades and never had a failure of any kind
even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip
blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this
volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got
bad/defective/BY chips..

Cheers

--
Raj, vu2zap
Bengaluru, South India.









--
  Michigan VHF Corp.   http://www.nobucks.net/   http://www.CDupe.com/
                          http://www.metalworkingfun.com




a.vision
 

Eh



Sent from my Samsung Galaxy smartphone.

-------- Original message --------
From: Gary Shriver <gshriver@...>
Date: 01/03/2018 17:13 (GMT+00:00)
To: BITX20@groups.io
Subject: Re: [BITX20] TDA2822 ill treatment test!

I’m no engineer here but could that be to the 2A current of your power supply?  Maybe the others are running a 20amp supply.  Just a thought..
73-Gary
> On Mar 1, 2018, at 4:04 AM, Raj vu2zap <rajendrakumargg@...> wrote:
>
>
> Having used TDA2822 for decades and never had a failure of any kind even with the chip running very hot,
> I got this urge to short my uBitx audio output and see it the chip blows.
>
> Power supply 13.6V
> Current limit 2A
>
> Touching the volume control leads produces a loud 50Hz hum. At this volume I directly shorted the wires
> at the EP jack. The chip got mildly warm thats it.
>
> No smoke or explosions ! I think that some may have got bad/defective/BY chips..
>
> Cheers
>
> --
> Raj, vu2zap
> Bengaluru, South India.
>
>
>
>
>




Dexter N Muir
 

p.s. another 'cheapo' move: no output caps. :)

Jerry Gaffke
 

Hmm, the TDA2822 does have a "bridge" mode where it is wired up with the two amps
out of phase to drive a speaker without needing output caps.
Is that what you are referring to?

The uBitx does not use bridge mode, the second section of the TDA2822 is left unused.
So the output load is between the amp output pin and ground,
allows the headphone jack to have the return at ground potential.
A DC blocking cap (C77 in the uBitx) is required.

> Shorting the output keeps both legs of it above ground.
Perhaps you mean that if we were in bridge mode with a 0 ohm load,
both sides of the load would wind up at a voltage above ground?

Jerry


On Thu, Mar 1, 2018 at 10:29 am, Dexter N Muir wrote:
Read the thread/s thoroughly, Raj (and all). Shorting the output keeps both legs of it above ground. It would seem that taking either leg to ground (that 'ring' connection that is taken there by a mono plug in particular) does the damage. It's the 'cheapo' method of boosting audio output: bi-phase two amps: 4x the power from the same supply voltage. This package evidently does this internally. If the spec sheet shows the internal circuit, this becomes obvious. I await reports...

 

No, the chip on short did not draw a large current.

Raj

At 01-03-2018, you wrote:
I suspect the 2A limit on his power supply saved the chip!

Ron W7HD


On 03/01/2018 10:00 AM, Joe Puma wrote:
Thanks for taking one for the team Raj. ðŸ‘ðŸ¼ðŸ‘ðŸ¼ðŸ‘ðŸ¼
On Mar 1, 2018, at 7:04 AM, Raj vu2zap <@Raj> wrote:


Having used TDA2822 for decades and never had a failure of any kind even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got bad/defective/BY chips..

Cheers

--
Raj, vu2zap
Bengaluru, South India.



--
W7HD - NAQCC#7587 OMISS#9898 KX3#6966 LinuxUser#415320


 

Dexter,

uBitx amp is single ended part of stereo amp. It is not bridged!
One lead is ground and other goes to output cap.

In bridge both outputs are used and cap may not be needed.

Raj


At 01-03-2018, you wrote:
Read the thread/s thoroughly, Raj (and all). Shorting the output keeps both legs of it above ground. It would seem that taking either leg to ground (that 'ring' connection that is taken there by a mono plug in particular) does the damage. It's the 'cheapo' method of boosting audio output: bi-phase two amps: 4x the power from the same supply voltage. This package evidently does this internally. If the spec sheet shows the internal circuit, this becomes obvious. I await reports...

 

Did some more tests.

Normal audio use the current was 60-70mA

Dead short with varied drive the current was 300-800mA. At 800mA the chip got hot!
The current did not go above 1A at any time.

I suspect there are TDA2822Ds masquerading as TDA2822M.

Raj

At 01-03-2018, you wrote:

Having used TDA2822 for decades and never had a failure of any kind even with the chip running very hot,
I got this urge to short my uBitx audio output and see it the chip blows.

Power supply 13.6V
Current limit 2A

Touching the volume control leads produces a loud 50Hz hum. At this volume I directly shorted the wires
at the EP jack. The chip got mildly warm thats it.

No smoke or explosions ! I think that some may have got bad/defective/BY chips..

Cheers

--
Raj, vu2zap
Bengaluru, South India.

Jerry Gaffke
 

Raj,

Good information, that 800ma when shorted.
There is likely a surge that is considerably higher while it charges the cap if you power up the rig
first with no headphones or speaker, then plug in a dead short.
Only there for maybe a couple milliseconds, so would need to use a scope to see it 
as a voltage across perhaps 0.1 ohm resistor that is in series with power to the chip.

Was that the stock TDA2822M that came with the uBitx?
Mine has a logo of "FCI" or maybe it's "FOI", where the middle character is filled in to where it is an almost solid circle.
It also says "PI1" and then below it says "TDA2822M".
I have no idea who makes that.

Post 43211 says there's some really bad clones masquerading as TDA2822M's.
There seems to be at least a half dozen different manufacturers of TDA2822 variants.
So any testing had best fully describe manufacturer and markings and such.

Tayda claims to have some  ST/SGS-Thomson TDA2822M's and TDA2822L's at cheap prices:
    https://www.taydaelectronics.com/catalogsearch/result/?q=tda2822
When you click through to the datasheets, you get either a 1995 ST datasheet (for the TDA2822M)
or a UTC datasheet (for the TDA2822L)    The "L" just means lead free.
A google search for TDA2822M finds this ST datasheet first, dated 2003 and marked "Obsolete":    http://www.st.com/content/ccc/resource/technical/document/datasheet/9e/18/f7/cd/2e/b8/43/62/CD00000134.pdf/files/CD00000134.pdf/jcr:content/translations/en.CD00000134.pdf  
I suspect the Tayda parts are not really ST. 

Yes, bridge mode would be nice for driving an internal speaker.
But not for driving anything external, were everyone would assume the jack shell is at ground potential.

Jerry



On Thu, Mar 1, 2018 at 07:22 pm, Raj vu2zap wrote:
Did some more tests.

Normal audio use the current was 60-70mA

Dead short with varied drive the current was 300-800mA. At 800mA the chip got hot!
The current did not go above 1A at any time.

I suspect there are TDA2822Ds masquerading as TDA2822M.

Raj
. . .

 

Emacs!


At 02-03-2018, you wrote:
Raj,

Good information, that 800ma when shorted.
There is likely a surge that is considerably higher while it charges the cap if you power up the rig
first with no headphones or speaker, then plug in a dead short.
Only there for maybe a couple milliseconds, so would need to use a scope to see it
as a voltage across perhaps 0.1 ohm resistor that is in series with power to the chip.

Was that the stock TDA2822M that came with the uBitx?
Mine has a logo of "FCI" or maybe it's "FOI", where the middle character is filled in to where it is an almost solid circle.
It also says "PI1" and then below it says "TDA2822M".
I have no idea who makes that.

Post 43211 says there's some really bad clones masquerading as TDA2822M's.
There seems to be at least a half dozen different manufacturers of TDA2822 variants.
So any testing had best fully describe manufacturer and markings and such.

Tayda claims to have some  ST/SGS-Thomson TDA2822M's and TDA2822L's at cheap prices:
    https://www.taydaelectronics.com/catalogsearch/result/?q=tda2822
When you click through to the datasheets, you get either a 1995 ST datasheet (for the TDA2822M)
or a UTC datasheet (for the TDA2822L)    The "L" just means lead free.
A google search for TDA2822M finds this ST datasheet first, dated 2003 and marked "Obsolete":    http://www.st.com/content/ccc/resource/technical/document/datasheet/9e/18/f7/cd/2e/b8/43/62/CD00000134.pdf/files/CD00000134.pdf/jcr:content/translations/en.CD00000134.pdf  
I suspect the Tayda parts are not really ST.

Yes, bridge mode would be nice for driving an internal speaker.
But not for driving anything external, were everyone would assume the jack shell is at ground potential.

Jerry



On Thu, Mar 1, 2018 at 07:22 pm, Raj vu2zap wrote:
Did some more tests.

Normal audio use the current was 60-70mA

Dead short with varied drive the current was 300-800mA. At 800mA the chip got hot!
The current did not go above 1A at any time.

I suspect there are TDA2822Ds masquerading as TDA2822M.

Raj
. . .

Mike Woods
 

That's what my one looks like in my initial batch uBITx.     It would be interesting to see photos of those that have had "the smoke come out".

This is the page that links from FCI to the datasheet:  http://www.alldatasheet.com/datasheet-pdf/pdf/210172/FCI/TDA2822M.html

And this shows a maximum voltage of 15v.

Mike ZL1AXG

On 2/03/18 5:38 PM, Raj vu2zap wrote:
Emacs!


At 02-03-2018, you wrote:
Raj,

Good information, that 800ma when shorted.
There is likely a surge that is considerably higher while it charges the cap if you power up the rig
first with no headphones or speaker, then plug in a dead short.
Only there for maybe a couple milliseconds, so would need to use a scope to see it
as a voltage across perhaps 0.1 ohm resistor that is in series with power to the chip.

Was that the stock TDA2822M that came with the uBitx?
Mine has a logo of "FCI" or maybe it's "FOI", where the middle character is filled in to where it is an almost solid circle.
It also says "PI1" and then below it says "TDA2822M".
I have no idea who makes that.

Post 43211 says there's some really bad clones masquerading as TDA2822M's.
There seems to be at least a half dozen different manufacturers of TDA2822 variants.
So any testing had best fully describe manufacturer and markings and such.

Tayda claims to have some  ST/SGS-Thomson TDA2822M's and TDA2822L's at cheap prices:
    https://www.taydaelectronics.com/catalogsearch/result/?q=tda2822
When you click through to the datasheets, you get either a 1995 ST datasheet (for the TDA2822M)
or a UTC datasheet (for the TDA2822L)    The "L" just means lead free.
A google search for TDA2822M finds this ST datasheet first, dated 2003 and marked "Obsolete":    http://www.st.com/content/ccc/resource/technical/document/datasheet/9e/18/f7/cd/2e/b8/43/62/CD00000134.pdf/files/CD00000134.pdf/jcr:content/translations/en.CD00000134.pdf  
I suspect the Tayda parts are not really ST.

Yes, bridge mode would be nice for driving an internal speaker.
But not for driving anything external, were everyone would assume the jack shell is at ground potential.

Jerry



On Thu, Mar 1, 2018 at 07:22 pm, Raj vu2zap wrote:
Did some more tests.

Normal audio use the current was 60-70mA

Dead short with varied drive the current was 300-800mA. At 800mA the chip got hot!
The current did not go above 1A at any time.

I suspect there are TDA2822Ds masquerading as TDA2822M.

Raj
. . .


--
Mike Woods
mhwoods@...

K5ESS
 

http://www.datasheetlib.com/datasheet/973902/tda7221a_fci-semiconductor.html  

FCI Semiconductors.   Link says TDA7221a as does page header  but datasheet is for TDA2822M.

Mike

K5ESS

 

From: BITX20@groups.io [mailto:BITX20@groups.io] On Behalf Of Raj vu2zap
Sent: Thursday, March 1, 2018 10:38 PM
To: BITX20@groups.io
Subject: Re: [BITX20] TDA2822 ill treatment test!

 

Emacs!


At 02-03-2018, you wrote:

Raj,

Good information, that 800ma when shorted.
There is likely a surge that is considerably higher while it charges the cap if you power up the rig
first with no headphones or speaker, then plug in a dead short.
Only there for maybe a couple milliseconds, so would need to use a scope to see it
as a voltage across perhaps 0.1 ohm resistor that is in series with power to the chip.

Was that the stock TDA2822M that came with the uBitx?
Mine has a logo of "FCI" or maybe it's "FOI", where the middle character is filled in to where it is an almost solid circle.
It also says "PI1" and then below it says "TDA2822M".
I have no idea who makes that.

Post 43211 says there's some really bad clones masquerading as TDA2822M's.
There seems to be at least a half dozen different manufacturers of TDA2822 variants.
So any testing had best fully describe manufacturer and markings and such.

Tayda claims to have some  ST/SGS-Thomson TDA2822M's and TDA2822L's at cheap prices:
    https://www.taydaelectronics.com/catalogsearch/result/?q=tda2822
When you click through to the datasheets, you get either a 1995 ST datasheet (for the TDA2822M)
or a UTC datasheet (for the TDA2822L)    The "L" just means lead free.
A google search for TDA2822M finds this ST datasheet first, dated 2003 and marked "Obsolete":    http://www.st.com/content/ccc/resource/technical/document/datasheet/9e/18/f7/cd/2e/b8/43/62/CD00000134.pdf/files/CD00000134.pdf/jcr:content/translations/en.CD00000134.pdf  
I suspect the Tayda parts are not really ST.

Yes, bridge mode would be nice for driving an internal speaker.
But not for driving anything external, were everyone would assume the jack shell is at ground potential.

Jerry



On Thu, Mar 1, 2018 at 07:22 pm, Raj vu2zap wrote:

Did some more tests.

Normal audio use the current was 60-70mA

Dead short with varied drive the current was 300-800mA. At 800mA the chip got hot!

The current did not go above 1A at any time.

I suspect there are TDA2822Ds masquerading as TDA2822M.

Raj

. . .

n5ib_2
 

Here's the one on mine.
852 of batch 2
With apologies to Monty Python.. it's not dead yet
And has a 10 ohm resistor...

N5IB

Vince Vielhaber
 

Better than Princess Bride: It's been mostly dead all day.

No, not referring to mine.
Vince.

On 03/02/2018 09:00 AM, n5ib_2 wrote:
Here's the one on mine.
852 of batch 2
With apologies to Monty Python.. it's not dead yet
And has a 10 ohm resistor...

N5IB