Audiodoc
Well-Known Member
OP is correctly placed. Possible my 200th IC insertion. Will check all the connections to check if there is some short circuit somewhere.
Group Buy: Mauro Penasa MyRef Rev C kit
- Thread starter linuxguru
- Start date
- Status
Audiodoc
Well-Known Member
Its a TF so no isolation pads required. Will check the resistors
Audiodoc
Well-Known Member
Here is the pic. Will just explain what happens when I turn on no sound. When I turn of the relay sounds and led blinks.
I can't find anything obviously wrong - everything seems to be correct on the boards. One thing that worries me is that the IEC power connector is very close to the heatsink, but that's not relevant to this problem. There's no short between the chassis, spacers and the boards, right?
The LM3886TFs look good from the top, but they could still be counterfeits/re-marks of higher quality - what's the batch number written on the face? One possible re-mark is an LM3876 that has been re-badged as an LM3886 in China, but there's no easy way to tell the difference. The LM3876 will have one +Vs power pin open (NC), while the LM3886 has both +Vs pins connected to the die.
Check if R11 (1 ohm) is open. If it's open, there is probably some further issue with the grounding or the rectifier/PSU section. If R11 is good (1 ohm), the next thing to try is to swap or replace the LM318 with good ones from known good sources. Debug just one channel first, after disconnecting the other board from the trafo. If changing the LM318 doesn't solve the problem, the LM3886TF is suspect.
The LM3886TFs look good from the top, but they could still be counterfeits/re-marks of higher quality - what's the batch number written on the face? One possible re-mark is an LM3876 that has been re-badged as an LM3886 in China, but there's no easy way to tell the difference. The LM3876 will have one +Vs power pin open (NC), while the LM3886 has both +Vs pins connected to the die.
Check if R11 (1 ohm) is open. If it's open, there is probably some further issue with the grounding or the rectifier/PSU section. If R11 is good (1 ohm), the next thing to try is to swap or replace the LM318 with good ones from known good sources. Debug just one channel first, after disconnecting the other board from the trafo. If changing the LM318 doesn't solve the problem, the LM3886TF is suspect.
Last edited:
Audiodoc
Well-Known Member
R11 is normal showing appx 1.4 ohms.
Voltage between + speaker out and pGND is 25 volts.
AC input is 25-0-25V
DC at smoothing Capacitors 32-0-32V
Input to LM318 is +-12V (24V).
resoldered the chip. Checked for any short circuits between the 3886 pins. Although from china the chips look original even the package is from National semiconductor. I have kept two more as a backup.
Have NE5532 too can they be used for checking.
Voltage between + speaker out and pGND is 25 volts.
AC input is 25-0-25V
DC at smoothing Capacitors 32-0-32V
Input to LM318 is +-12V (24V).
resoldered the chip. Checked for any short circuits between the 3886 pins. Although from china the chips look original even the package is from National semiconductor. I have kept two more as a backup.
Have NE5532 too can they be used for checking.
Audiodoc
Well-Known Member
I check for some troubleshooting on diyaudio. This is what I got: My problem is similar the relays not closing. Nothings burnt / smoked. Should I replace a 3886.
"The active componets are fairly robust. The LM318 is easier to kill than the LM3886. You have to work hard to kill it.
After checking all voltages, the offset is usually a good indicator of where to go. High offset and no relay go together. Offset is either a problem with rails or the feedback loop.
An open in the feedback loop will give high offsets and prevent the realy from closing.
Aftter that look and smell for toasted resistors. It also is possible to smoke the zeners, and the 1K dropping resistors, small diodes feeding the relay/led circuit, and other necessary parts.
But always start at the rails for the LM3886, AND LM318. Then output dc offset. This should ID 75% of any issues.
I am running in a new one using the mono boards now. Hope to post pics tonight. Love all three listened to so far."
"The active componets are fairly robust. The LM318 is easier to kill than the LM3886. You have to work hard to kill it.
After checking all voltages, the offset is usually a good indicator of where to go. High offset and no relay go together. Offset is either a problem with rails or the feedback loop.
An open in the feedback loop will give high offsets and prevent the realy from closing.
Aftter that look and smell for toasted resistors. It also is possible to smoke the zeners, and the 1K dropping resistors, small diodes feeding the relay/led circuit, and other necessary parts.
But always start at the rails for the LM3886, AND LM318. Then output dc offset. This should ID 75% of any issues.
I am running in a new one using the mono boards now. Hope to post pics tonight. Love all three listened to so far."
PSU and protection sections are fine, rails are fine, and there are no rail shorts.
You can use a DIP8 NE5534 (in place of LM318) to test - it will work, and sounds pretty good actually. OPA627A and LME49710 will also work, with different levels of stability. In a pinch, TL071 may work for testing (haven't tried it, but it works in simulation).
For the feedback loop(s), check the 470k, 12k and 390R resistors. Then C9 (220uF BG-PK) for polarity and shorts. Also re-check the 22pF blue C0G ceramic caps for a short - these can be replaced by anything up to about 100pF, and will give more stability with higher values.
Last edited:
Audiodoc
Well-Known Member
What should be the DC voltage between the +speaker out and PGND like.
Audiodoc
Well-Known Member
After I removed the LM318 from the IC base and turned on the amplifier the relay normally clicked and LED lit. So will buy new 318's tomorrow as I have only 1 TL071, 2 NE5532 (which are dual opamps and wont work), 3 OPA2134 which I have used to mod my Norge (Dont want to maim a working amplifier) and two OPA627 which are costly. If I dont get one will have to trouble you.
Last edited:
DC voltage between speaker out and PGND should be below 0.1V (100 mV) in normal operation, with no signal at the input. In most cases, it is below 10 mV.
If the TL071 is a DIP8, you can try it out. It's inexpensive, so nothing to lose if it smokes (which isn't likely, because the opamp rail voltages look safe). No point in risking an OPA627, of course. NE5534A is a reasonable ~Rs.20 possibility for testing if the LM318 isn't available. Feel free to ping me if none of these are available.
There was an earlier case of a NatSemi LM318 causing a high latch-up at the output, but this is the first TI LM318 with the same issue. If they're available, try to get a pair of NatSemi LM318N as well as TI LM318P/J as the alternate.
Last edited:
Audiodoc
Well-Known Member
Put the TL071 with fingers crossed and pushed to power switch. And voila everything seems fine. Will do a speaker test later. Why should both LM318s be bad even when everything was fine.
Audiodoc
Well-Known Member
Now the voltage between the speaker out and power ground is between 0.001 - 0.002V.
Does this mean I cant use the 318s in other projects too.
Does this mean I cant use the 318s in other projects too.
Cool - I hope you get sound, at least from one channel if not both.
Bad LM318s: Not sure - Uriah reported on diyaudio that a few of his NatSemi LM318s had the same issues on some boards, that were resolved by replacement. The bad LM318s worked without latch-up on some other boards, it was probably a marginal input offset current/voltage issue.
Maybe ESD? It's possible that North India is much drier now, than B'lore, allowing static to build up and damage the inputs (which have back-to-back protection diodes, BTW). The other possibility is X-Ray damage to Air Cargo from scanners.
Offset voltage: Very good - among the best so far. But it may not be this good when the JFET-input TL071 is replaced by a good LM318 eventually.
LM318: It may work just fine even in another MyRef board, and most probably in other projects that require a uncompensated single-opamp. AFAIK, that latch-up issue hasn't been fully investigated or solved yet.
Audiodoc
Well-Known Member
Yes I got sound in the single monoblock I tested tonight will try to find the opamps tomorrow. If I dont get them will trouble you.
Had purchased the 3886 pretty cheap in an ebay auction 7 for 1200 bucks rest of the amplifier based on the chips were working fine didnt want to buy from element 14 (costs 420 and takes 2 weeks to arrive).
Had purchased the 3886 pretty cheap in an ebay auction 7 for 1200 bucks rest of the amplifier based on the chips were working fine didnt want to buy from element 14 (costs 420 and takes 2 weeks to arrive).
Last edited:
Audiodoc
Well-Known Member
Searched every shop in Chandigarh, none had the LM318. Although I have checked both the channels upon inserting the the TL071 and they sound fine. One more thing, I have a 50K blue velvet Alps pot but it causes motorboating will try a shunt mod tonight. I know 10-20k log would have been a preferred pot but I only have 50K ones.
What should be the rating for the power supply fuse for MyRef Rev C?
Update: did get the cab from Dinrack. This is raw, unpainted, no holes. The panels are rather flimsy - OK, make that quite flimsy - at 2.5mm (I think). Good thing, though, is it has two layers in the bottom. Upper bottom layer is where one mounts the circuit board and other stuff. Lower bottom plate is the actual cover. Dinrack has made a nice punch to fit the toroidal transformer's mounting plate.
Yet to find the time to sit down, measure dimensions, and eventually make proper holes for fitting various connectors. Need to go back to Dinrack after making holes for getting it painted.
Also got the heat sinks (Visha Electronics).
Update: did get the cab from Dinrack. This is raw, unpainted, no holes. The panels are rather flimsy - OK, make that quite flimsy - at 2.5mm (I think). Good thing, though, is it has two layers in the bottom. Upper bottom layer is where one mounts the circuit board and other stuff. Lower bottom plate is the actual cover. Dinrack has made a nice punch to fit the toroidal transformer's mounting plate.
Yet to find the time to sit down, measure dimensions, and eventually make proper holes for fitting various connectors. Need to go back to Dinrack after making holes for getting it painted.
Also got the heat sinks (Visha Electronics).
Audiodoc
Well-Known Member
I got one from hongkong, these are better than Dinrack but costly. 3mm throughout 8mm front panel. Since my project was already crossing 10k went a step further to make it look that costly and compact.
Had got is ready a day before but the wretched 318s went mad! So everything was disassembled. Waiting for some tamed 318s from linuxguru.
Had got is ready a day before but the wretched 318s went mad! So everything was disassembled. Waiting for some tamed 318s from linuxguru.
Audiodoc
Well-Known Member
Tried again didn't find any lm318s. Bought a pair of NE5534s for testing but the DC at the output is in the higher range. 200-400mV as compared to the JFET TL071 where it is in the 0.001 range.
- Status
Follow HiFiMART on Instagram for offers, deals and FREE giveaways!
