Gearslutz

Gearslutz (https://www.gearslutz.com/board/)
-   Geekslutz Forum (https://www.gearslutz.com/board/geekslutz-forum/)
-   -   Summit TLA 100A opamp failure (https://www.gearslutz.com/board/geekslutz-forum/1292456-summit-tla-100a-opamp-failure.html)

Radardoug 26th December 2019 12:45 AM

Summit TLA 100A opamp failure
 
2 Attachment(s)
So I've had a Summit TLA 100A on the bench for service. Apparently the bright spark who designed this was not that bright. There is a Bal/Unbal switch on the back panel. In the Unbal position, one of the John Hardy copy opamps is shorted to ground via a 75 ohm resistor. Now Johns opamps are specced for working into 75 ohms continuously, but apparently the Summit copies are not. So the opamp dies from heat exhaustion. Now the idea of shorting an output to set up an operating mode is just plain dumb. What needs to happen is that the output pin on the XLR should be grounded, and the opamp disconnected from the output pin. Fortunately this mod is possible. Follow these steps.

1/ Remove the brown wire from the Bal/Unbal switch from point D on the pcb.
2/ Cut the track from the bypass relay that connects to trace D. Cut close to
the relay so you only cut the relay connection off.
3/ Connect brown wire to this disconnected relay contact.
4/ Connect a wire from the unused contact on the Bal/Unbal switch to D on the pcb.

I've attached a couple of pictures.

Note that when the opamp dies, the whole output is affected, because the bright spark connected the two opamps in series, and death of one affects the other. It also puts full rail d.c. on your output! Well done bright spark! Feel free to justify your design decisions if you read this.

As a side note, you can get replacement opamps of higher quality and cheaper from the great John Hardy, still alive and doing good work. Please support him rather than Summit, who ripped him off.

cathode 26th December 2019 07:06 AM

Summit SA990
 
1 Attachment(s)
what Summit suggests:

Radardoug 26th December 2019 08:05 PM

Yes I guess they would say that, having made a design which is sure to cook up one of the opamps! How to make a continuing business! And yet, they still went broke. My philosophy is you build equipment to not fail.

Clive Banks 28th December 2019 10:21 AM

I have serviced Summit equipment in the past . Thank you for this information.

John Hardy 28th December 2019 03:29 PM

If the Summit SA990 op-amp is so different that the 990s that I (and others) make are "incompatible" with Summit equipment, then perhaps Summit should not be calling their op-amp a "990" at all. Seems to me it is their op-amp that is incompatible.

But of course, it is BS on their part. And Summit still owes me $2000 from years ago, when I sold them gold-plated pins and black-anodized aluminum potting shells to help them out. As the saying goes, "No good deed goes unpunished."

I have sold many of my 990 op-amps to Summit owners and repair people to replace fried Summit SA990 op-amps.

My 990 web page:

http://www.johnhardyco.com/990OpAmpDetails.html

Thank you very much.

John Hardy
The John Hardy Co.
http://www.johnhardyco.com/index.html

VertigoRecording 28th December 2019 04:35 PM

What John said. I have fixed quite a few of these Summit units using Hardy 990s. Total BS on their part regarding the "special" op amp they utilize. - Charlie Bolois @ Vertigo

samwinston123 30th December 2019 02:18 AM

Quote:

Originally Posted by Radardoug (Post 14405686)
So I've had a Summit TLA 100A on the bench for service. Apparently the bright spark who designed this was not that bright. There is a Bal/Unbal switch on the back panel. In the Unbal position, one of the John Hardy copy opamps is shorted to ground via a 75 ohm resistor. Now Johns opamps are specced for working into 75 ohms continuously, but apparently the Summit copies are not. So the opamp dies from heat exhaustion. Now the idea of shorting an output to set up an operating mode is just plain dumb. What needs to happen is that the output pin on the XLR should be grounded, and the opamp disconnected from the output pin. Fortunately this mod is possible. Follow these steps.

1/ Remove the brown wire from the Bal/Unbal switch from point D on the pcb.
2/ Cut the track from the bypass relay that connects to trace D. Cut close to
the relay so you only cut the relay connection off.
3/ Connect brown wire to this disconnected relay contact.
4/ Connect a wire from the unused contact on the Bal/Unbal switch to D on the pcb.

I've attached a couple of pictures.

Note that when the opamp dies, the whole output is affected, because the bright spark connected the two opamps in series, and death of one affects the other. It also puts full rail d.c. on your output! Well done bright spark! Feel free to justify your design decisions if you read this.

As a side note, you can get replacement opamps of higher quality and cheaper from the great John Hardy, still alive and doing good work. Please support him rather than Summit, who ripped him off.

Hmm, do you have a schematic for the output stage? Are you sure it isn't a cross coupled / "quasi-floating" output? 75 ohms is common for the output sensing resistor, which is shorted to ground in order to double the output level on the other signal line for an unbalanced connection.

http://www.douglas-self.com/ampins/balanced/balfig6.gif

Radardoug 31st December 2019 03:46 AM

No, its not cross coupled, just two brute force inverting stages in series. When the opamp does die, the other opamp stops working as it is fed from the first. They both have d.c. servos as well.