speaker protection relay issue

[Enzo]

OK, they replaced the IC with the little board. But look at the schematic, all four caps ar still there. C315,316,317,318. The circuit is the same as the older style other than the chip is made up from parts. You could remove that little board and solder in a real TA7317 and it would never know the difference.

[pontiacpete]

Daz, I had a hartke here with that 1x1 pcb with surface mounted components in place of the TA7317. And Hartke sent the TA7317 for a replacement as in JPs picture.

[daz]

So i assume you guys are saying you think it IS the protect module/IC ? By the way, Jazz said 8 pin...why? The 7313 shows 9 and the module on this one is 9. pete....did you have the same issue on the one you fixed where it would trip the relay sometimes within seconds other times not at all?

[pontiacpete]

No, it was a no sound problem.

BTW I have 2 of them here if you need one.

[daz]

No, it was a no sound problem.

BTW I have 2 of them here if you need one.

Thanks !!! I appreciate that very much. But i'm sure i can get one. If not i'd contact you but with all the online sources it'll be easy to find. thanks. But before i do, anyone know how i can determine with some degree of surety that it is indeed the IC?

[daz]

And by the way, thanks to everyone here. The reason i ask questions like this here is because there are a number of super helpful and very talented techs here, and i just wanted to say i really appreciate your efforts to help people the way you do. And not to belittle any of you in the least because you are all extremely helpful and very knowledgeable, but i have to especially give a thumbs up to Enzo because not only is he as expert as expert gets, but i have seen him helping people at several sites for quite a few years tirelessly. He's gotta have a million posts trying to help those like me who are sub moronic where electronics theory is concerned. A special thanks to him.

[daz]

I may have a hint here for you all. It's beyond me but you may be able to come to some sort of conclusion from it. I was reading a post i found elsewhere after googling TA7317 where a guy was having the same issue as me tho a totally different amp with the same protection IC. he said he finally found it was a shorted diode or something along thos lines and mentioned it had to do with pin 1. Something about the pin being grounded but wasn't when it should have been. Seems thats the opposite of what a shorted diode would cause, but in any case i checked and pin one is 100% open in respect to ground with the amp off. When i turn the amp on it immediately goes to zero resistance even before the realy clicks on. I then did what he said and connected pin one to ground with the amp on using a wire and tried this a couple times where the relay either clicked on or off. It caused it to trigger at time where normally it wouldn't like when the amp is working correctly for a few minutes. If it stays on for more than about 30 second it will not click off till i turn the amp off and back on several times and finally i'll get it to go into protect. But touching pin 1 to ground triggered it one time.

I have no idea what all that points to, i'm just reporting some thing i found that i thought might help make sense of this to some of you.

[Jazz P Bass]

If you want to repair this amp, you need data.
IF the ic is bad can only be determined by looking at all of the pins & documenting what is measured.
This is a fairly complex circuit, whether it is the IC or the discrete circuit on it's own board.
The pins on the header are the same pins as the IC.
Measure those voltages.
Then it can be ascertained if the problem is the ic or not.
Disabling the circuit may tell you other bits of information.
I still would like to see how far the offset voltage goes, without the protection hooked up.
Information is king.
And it has firm roots.
Without it you will just be bouncing around.

[daz]

If you want to repair this amp, you need data.
IF the ic is bad can only be determined by looking at all of the pins & documenting what is measured.
This is a fairly complex circuit, whether it is the IC or the discrete circuit on it's own board.
The pins on the header are the same pins as the IC.
Measure those voltages.
Then it can be ascertained if the problem is the ic or not.
Disabling the circuit may tell you other bits of information.
I still would like to see how far the offset voltage goes, without the protection hooked up.
Information is king.
And it has firm roots.
Without it you will just be bouncing around.

Will do. I'll measure all 9 pins while the amp is functioning then again when it's in protect mode and post them. Might not get to it till tomorrow mid day but i will definately do it. Thanks.

[daz]

Ok, here are the voltages. One notable thing is pin 6 is at +97v when the relay is off and down to nothing when on.

Voltages when relay is off (protect on)
1)-8v
2).450mV
3)215mV
4)0
5) -96v
6)+97v
7)15mV
8).800mV
9)3.1v

Voltages when protect is off (relay on, amp working)
1)-8v
2).450mV
3)215mV
4)0
5) -96v
6)40mV
7)0
8)1.5v
9)3.1v

[Jazz P Bass]

Ok, here are the voltages. One notable thing is pin 6 is at +97v when the relay is off and down to nothing when on.

The purpose of the protect circuit is to give or not give the relay 'ground'.
That is what you see at pin 6.
When Pin 8 goes to 1.5v's, that turns on Q2.
Q2 turns on Q3, whose collector is the relay voltage.(pin 6)
When Q3 turns on, it gives the relay coil 'ground', sinking pin 6's voltage.
The listed voltages appear normal.
As designed, there is a slight delay at power up before the relay engages.
That is Pin #8 charging up C316.
So this is how the amp should work.
And it appears that it is functioning.
What pin voltage changes when it just up & turns off for no apparent reason?

[Jazz P Bass]

Thinking further: Why is pin1 so high? (8v's)
(The TA7317 datsheet indicates 0.75 v's)
You would think that would turn on Q1, which should totally kill Q2 base voltage (pin8)

[daz]

Thinking further: Why is pin1 so high? (8v's)
(The TA7317 datsheet indicates 0.75 v's)
You would think that would turn on Q1, which should totally kill Q2 base voltage (pin8)

Well, for one thing i made a mistake in my readings in the above post....pin one is -8v, not +. I will change that in the post now.
Also, pin 2 starts off when i turn the amp on around 100mV or thereabouts and gradually creeps up to somwhere around 450mV then either sits there and the amp continues to work, or it will go into protect and the voltage stays at about 1/2 a volt. But it constantly varies slightly which i guess if the PSU fluctuating for some reason. So maybe it's flaky cap in the psu? Ahh....and as i'm writing this it went out of protect and i checked the pin 2 voltage and it should about 300mV. So maybe it's the psu fluctuating back and fourth crossing the line that triggers it? I suppose i should check the psu next, yes?

[Enzo]

Don't worry about the 97v on the relay. When the relay is off, no current flows, so there is zero voltage drop from whatever supply it connects to. There is a resistor in series. When the IC grounds that point, the resistor limits current through the relay coil. The relay coil may have a voltage raqting, but it will bhe current that burns it out, if ever. That current is limited. The 97v is not wrong.


So pin 1 is -8v. Scope it, or flip your meter to AC volts. Is it clean -8v or is there ripple there? C314 and D309 form a small power supply just for that. A bad cap leaves ripple, which could put the circuit on edge.

[Enzo]

Don't worry about the 97v on the relay. When the relay is off, no current flows, so there is zero voltage drop from whatever supply it connects to. There is a resistor in series. When the IC grounds that point, the resistor limits current through the relay coil. The relay coil may have a voltage raqting, but it will bhe current that burns it out, if ever. That current is limited. The 97v is not wrong.


So pin 1 is -8v. Scope it, or flip your meter to AC volts. Is it clean -8v or is there ripple there? C314 and D309 form a small power supply just for that. A bad cap leaves ripple, which could put the circuit on edge.