I swapped about a dozen tl072 for lm4562 in my 30+ year old preamp.
After reading and reading and re-reading the service manual and opamps datasheets, I concluded I could just swap them. Now I've read in some other threads about decoupling caps needed.
It works and sounds fine after the swap.
So my question is, am I ok and safe (literally) electrically?
-david
After reading and reading and re-reading the service manual and opamps datasheets, I concluded I could just swap them. Now I've read in some other threads about decoupling caps needed.
It works and sounds fine after the swap.
So my question is, am I ok and safe (literally) electrically?
-david
The only way to know for sure is to test with an O'scope (recommend at least 20 MHz) or an analyzer, anything else is just guessing and hoping. Safety should be of little concern since it's working. Only caveat might be the added current draw of the replacement chips, so it wouldn't hurt to monitor the temperature of the power supply components. If you don't have access to the proper tools, you could also just make sure the decoupling scheme follows recommendations as closely as possible, it might provide you a little peace of mind.
Mike
Mike
Last edited:
You are safe... the thing I would want to check is whether the power supply is up to running a dozen LM4562's as they are more power hungry than TL072. You really do need to look at that aspect. See how they are fed, and if its a series regulator then check its not running to hot and that there is sufficient voltage for the reg to work with. If the feed were resistive plus zeners then almost certainly the opamps would be running on reduced voltage.
Check it !
As to decoupling, I would be amazed if caps were not already present for the TL072's.
Edit... I see Michael had the same thoughts on the power requirements
This is great! Just the advice I was looking for.
I can run it for hours at a time with no adverse effects. I will open it up next chance to see how hot they are and what voltage they see.
The opamp's power supply takes +/- 24 volt off of the rectifier, through some 7818/7918 to regulate to +/- 18 volt, then through a diode string to +/-16. I think the servivce manual notes state that 16vdc is the max rating. It's powering 11 lm4562s
I could always tinker with the ps or even replace it, woukd be a fun project.
I can run it for hours at a time with no adverse effects. I will open it up next chance to see how hot they are and what voltage they see.
The opamp's power supply takes +/- 24 volt off of the rectifier, through some 7818/7918 to regulate to +/- 18 volt, then through a diode string to +/-16. I think the servivce manual notes state that 16vdc is the max rating. It's powering 11 lm4562s
I could always tinker with the ps or even replace it, woukd be a fun project.
78 and 79 regs will have no problem with the current as they are 1 amp devices but are they on a heatsink ? Eleven 4562's at say 12ma each would be 130ma or so. Lets say 150ma to cover other stuff as well (and it could be a bit more). 24 volts in, 18 out, that means the regs see at least 6 volts across them which at 150ma is nearly 1 watt. Doesn't sound much but I bet they are pretty hot to touch if not heatsinked, and if they are not then some clip on T0220 style fins would be advised.
The regulators are attached to part of the chassis acting as a heatsink. The regulators and 4562s are cool to the touch. So cool you wouldn't even know the preamp was on. I get approx 31.8v across the V+ and V- on the 4562s, so the supply must be doing its job correctly.
Unfortunately desoldering just one pin, and keeping it in good enough shape to resolder, would be risky for me. Are you looking to verify the current? I do have a few extra 4562s and will risk it if the test is important enough.
You may want to check if capacitors are used at the inputs of the opamps. LM4562 is a BJT input type opamp so it must have DC blocking caps at its input in almost any case. If electrolytic caps are used you could change those for film types in 2.5 mm or 5 mm pitch as those exist now. The preamp will sound better with film caps in the signal path. Also if output caps are used film caps will win hands down from electrolytic caps at that spot. Values like 4.7 µF can be found in MKT quality and 5 mm pitch. If you are unsure about decoupling please check if the decoupling caps are close to each (!) chip. Fast opamps like 100 nF from + pin to GND and one 100 nF from - pin to GND as close to each chip as possible. Wima MKS2 series exist in 2.5 mm pitch and they fit physically in many cases. This is when no decoupling caps are used, when let's say 4.7 µF caps are used you can leave those or replace them by new caps of good quality. I suppose you already changed all electrolytic caps as the preamp is 30 years old. I would replace all electrolytic caps because of age, as usual Panasonic FC is good for this purpose.
You could also use 1 piece 100 nF Wima MKS2 100 nF in 2.5 mm pitch from + to - at each chip if space is tight. At the bottom of the PCB straight from + to - pin.
*BTW LM4562 can be used with +/- 17V so no worries there. +/-18V are absolute maximum.
You could also use 1 piece 100 nF Wima MKS2 100 nF in 2.5 mm pitch from + to - at each chip if space is tight. At the bottom of the PCB straight from + to - pin.
*BTW LM4562 can be used with +/- 17V so no worries there. +/-18V are absolute maximum.
Last edited:
From the schematic it doesn't look like I have caps going to ground on the inputs. Also, I'm assuming you've seen boards from the 70s - traces are like meandering streams - no nice sharp angles, trace size changes to fit the space available, etc. Finding a close ground trace may be impissible. So a cap between the + and - inputs may be the best solution.
Just for kicks, I've attached a portion of the schematic and pcb layout (it's all available publicly so no concerns there).
The 4562's should certainly be detectably warm so if they are not then I would definitely try and measure the current to one of them. Solder braid and hot iron should cleanly remove solder from the pins, so cleanly that they should drop out of the board and be indistinguishable from new.... desoldering is a skill that has to be learned though.
As the original circuit used 4558's I don't foresee any issues... I'm sure the unit will be 100% AC coupled anyway.
As the original circuit used 4558's I don't foresee any issues... I'm sure the unit will be 100% AC coupled anyway.
Missed this comment - parts are from digikey so I'm not concerned about fakes.
Ah - so I desolder all pins, bend pin 8 out, solder the other 7 back in, attach resist between pin 8 and hole, measure. That sounds feasible for me.
I can try this too if you explain what pins to probe and what to look for. My oscope is very basic analog 20mhz but is dual trace.
Please don't do so just to be sure there is current. We know there is current flowing. Just decouple them an enjoy the device.
The pdf you posted dis not complete but I see there is decoupling but not with all opamps like IC8 (if that is an opamp, can't read it). If you have a full copy better comments could be given.
The pdf you posted dis not complete but I see there is decoupling but not with all opamps like IC8 (if that is an opamp, can't read it). If you have a full copy better comments could be given.
Last edited:
IC8 is an opamp. Just about every IC is an opamp.
The schematic should be complete. I know it is difficult to read the details, but that's all the Web has. BUT Apt did something strange and put a physical copy on the bottom of the unit. I'm trying to get some hi-rez photos of it.
The board shown isn't complete, but it is the main board. There are 2 others, but I don't have similar depictions. One is just for tape and other inputs/outputs - nothing interesting there. The other board has most of the switches - tape dubbing, tone defeat., some filters...and other parts of the schematic
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.