This receiver is fairly highly regarded, it was the largest to come out of the receiver wars of the 70s, making 330w per channel. Pretty outstanding if it isn’t the prettiest girl at the dance.
It uses 6 TA7136P seven pin single opamps. Now this opamp isn’t highly regarded, not even slightly. It’s known to be a pretty poor performer. I see that there are a couple of people that sell adapter boards to where you can attach your own opamp, but at $5-$7 per adapter board I’m going to look at other options. I am currently trying to get accustomed to Kicad and EasyEDA so that I can make up my own adapter boards and order 50 for less than the cost of the 6 I would have to buy (I will share the gerber when I do).
One board is supposed to be used with the OPA604, but it’s generic pinout where they just attach the appropriate pins, V+ to V+, output to output, etc. So this board can be used with any single opamp with the same pinout like the TL071. Speaking of, I’ve also the TA7136 repalced with a TL071.
The other board that is available can be used with single opamps like the OPA604 and NE5534AP, but can also be used with dual opamps like the TL072, NE5532AP. With the dual opamps you’re just supposed to wire the two opamps in parallel, so parallel the outputs, + and - inputs and such. I still don’t know what the advantage of using a dual opamp is over a single opamp. Is wiring a dual opamp in parallel give better performance?
Until I get well enough acquainted with Kicad or EasyEDA I’m thinking I’ll just use these small adapter boards I have on hand and wire leads that go to the appropriate spots on the TA7136 pinout.
I can use standard 8 pin opamps or I could use a smaller SOIC-8 package. For standard packages I have TL072CP, NE5532AP and then the single NE5534AP. For the smaller packages I have some OPA1656IDR and then for a single option I have OPA1641AID.
Im thinking for simplicity I’ll use of the single opamp options, unless someone has a reason as to why I should go dual.
It should fairly simple, joining the correct pins.
Just omit pins 1 and 5 and remove the components on the board at 1 and 5 as stated in the photo. Also attach a 100 nF from pins 4 to pin 7, the supplies.
So my questions, any reason to use a dual opamp over a single? And do all of my choices look like they would work well? That being NE5524 and OPA1641 for the singles and NE5532, OPA1656, or TL072 for the duals?
Appreciate any and all thoughts on this.
Dan
It uses 6 TA7136P seven pin single opamps. Now this opamp isn’t highly regarded, not even slightly. It’s known to be a pretty poor performer. I see that there are a couple of people that sell adapter boards to where you can attach your own opamp, but at $5-$7 per adapter board I’m going to look at other options. I am currently trying to get accustomed to Kicad and EasyEDA so that I can make up my own adapter boards and order 50 for less than the cost of the 6 I would have to buy (I will share the gerber when I do).
One board is supposed to be used with the OPA604, but it’s generic pinout where they just attach the appropriate pins, V+ to V+, output to output, etc. So this board can be used with any single opamp with the same pinout like the TL071. Speaking of, I’ve also the TA7136 repalced with a TL071.
The other board that is available can be used with single opamps like the OPA604 and NE5534AP, but can also be used with dual opamps like the TL072, NE5532AP. With the dual opamps you’re just supposed to wire the two opamps in parallel, so parallel the outputs, + and - inputs and such. I still don’t know what the advantage of using a dual opamp is over a single opamp. Is wiring a dual opamp in parallel give better performance?
Until I get well enough acquainted with Kicad or EasyEDA I’m thinking I’ll just use these small adapter boards I have on hand and wire leads that go to the appropriate spots on the TA7136 pinout.
I can use standard 8 pin opamps or I could use a smaller SOIC-8 package. For standard packages I have TL072CP, NE5532AP and then the single NE5534AP. For the smaller packages I have some OPA1656IDR and then for a single option I have OPA1641AID.
Im thinking for simplicity I’ll use of the single opamp options, unless someone has a reason as to why I should go dual.
It should fairly simple, joining the correct pins.
Just omit pins 1 and 5 and remove the components on the board at 1 and 5 as stated in the photo. Also attach a 100 nF from pins 4 to pin 7, the supplies.
So my questions, any reason to use a dual opamp over a single? And do all of my choices look like they would work well? That being NE5524 and OPA1641 for the singles and NE5532, OPA1656, or TL072 for the duals?
Appreciate any and all thoughts on this.
Dan
I guess another question I might have is of those options, which would be the best performer? I’ve read that the NE5534/32 is supposed to be a fantastic performer, especially at its price point. I don’t know how any of my options compared to the OPA604. The people that have replaced the originals with the TL071 seemed happy with the results so I’m guessing that any of my options would be certainly better than the originals.
Dan
Dan
A dual op amp has two op amps inside and you need a single op amp.
Never wire a TL072 or any dual op amp as a TL071 or any single op amp.
The uA741 would work ideally in this scenario as a slow op amp is required.
The TL range are FET input, not needed. Schematic of the original attached.
Never wire a TL072 or any dual op amp as a TL071 or any single op amp.
The uA741 would work ideally in this scenario as a slow op amp is required.
The TL range are FET input, not needed. Schematic of the original attached.
Attachments
I appreciate the response, I do realize that I am replacing a single op amp and I do know that dual op amps have two op amps in them, hence the name.
I’m not sure why you said to never wire them like that, as there are people that are indeed bridging the two opamps together. Here on this board they are bridging the two + inputs together, the two - inputs together.
And on the back of this same board, you’ll see there are two small pads, you solder those two pads together in order to bridge the two outputs together, so you could use the outputs individually or bridged.
So as you can see, you can certainly use dual opamps to replace this single opamp. Again, I don’t know the benefit of using a dual.
And you say not to use any TL as they are FET, again something that I did realize, but the TL071 has been used on multiple occasions for this replacement. Plus the popular one to use is the OPA604, isn’t that a FET input? Same with the OPA2134 which is what was recommend by the designer of that one adapter board. I did have the schematic of the original device as well, but I appreciate you posting it.
Any reason you would say to never bridge the two opamps or use FET opamps? I am genuinely asking, is there a reason not to?
Dan
Last edited:
If the amplifiers are not precisely matched, they will 'fight' for ouput drive current.
If they are not precisely balanced, and have a DC offset, that will cause current drain and sometimes distortion issues.
I suggest the only reason they used dual op amps is because the had run out of single op amps. Nothing more.
Use dual op amps if you want, there is no advantage.
If they are not precisely balanced, and have a DC offset, that will cause current drain and sometimes distortion issues.
I suggest the only reason they used dual op amps is because the had run out of single op amps. Nothing more.
Use dual op amps if you want, there is no advantage.
Thank you, I figured there was no advantage to it. Are the two op amps inside of a single dual op amp, are they fairly well matched?
In the end, it doesn’t really matter because I took a look at the supply voltages and the Technics and it’s +/-20v, so that basically eliminates most of what I was looking at. Most of those are 15 to 18 V at max. So I ordered a bunch of OPA604, it’s considered a nice and quiet device and it handles 24 V supplies. I appreciate you answering my question about the dual, I didn’t know if there was anything special, but you cemented my suspicions, I do appreciate it!
Dan
TA7136p are actually quite ok, raising the bias affects speed. These were in Sx-780 and they have a very low noise figure. 5534 may be similar noise.
I replaced them with 2 6922 tubes.
If you have a 100khz function generator you can see where slew distortion happens.
I replaced them with 2 6922 tubes.
If you have a 100khz function generator you can see where slew distortion happens.
Last edited:
Interesting, most places I read on them say that they’re actually quite poor and that replacing them with modern opamps (specifically the OPA604) reduced the noise floor greatly. I have a SX-780, I’ll look into that.
How do you replace them with tubes at those lower voltages?
Dan