Hi Owen,Superb job on building the LPUHP! I also appreciate the feedback on your listening experience
I like your mounting arrangement, and the modular look for the two-way setup. My guess is that you probably do not need heatsinks, but it really depends on your usage and listening levels / speaker efficiency.
The one driving the tweeters will almost certainly not need a heatsink, but the one running the midbass / bass might need one if you're listening at high levels for long periods of time.
Do you have any way of measuring the temp on the LME49610 during operation?
Regards,
Owen
I'm using your LPUHP amp in an active 3-way setup so the midrange has a high pass at 300hz. The amp doesn't get warm at all, I didn't measured the temperature but after 2 hours of listening I can leave my finger on the lme46100's.
The sound is exactly what I was looking for!!! Thanks again for this great design.
I'm waiting for my new amp that will drive my woofers to put everything in a nice case, I'll show some pictures when done to boost some lazy people to mount their precious
Alexandre
Here is a picture of my build (sorry for the bad image quality). When I powered it up for the first time the right channel had a significant (185mV) DC offset. While I was debugging the issue, something on the board (around the buffers I think) went ZAP (a loud click and a flash, but no smoke). At that point I thought that the amp was toasted for sure. However, after resetting the power the board seemed to be alive. At least I could read the DC offset again. In the end it turned out that one of the volume pot solder joints came off while I was mounting it and the right channel was hanging only by the 1M resistor, hence the DC offset. After a bit of soldering the offset was gone and I was able to plug my phones in and the music was flowing through nicely. Now I'm a little paranoid about whether that zap has damaged any opamps in some subtle way.
Hi people,
after I couldn't find a solution how to connect both LPUHPs to one transformer, I decided to buy new regulators, as they maybe got damaged by connecting bith rails to one secondary.
As I already wrote the amps do although function, but I wonder about the temperature of the heatsink, which is getting warm after about 5 minutes with only really a little power.
I do also wonder why I have different results for the voltage between A, B and C, as Pwen wrote some posts ago. I hope by changing the regs all problems are gone.
Another question regarding the LME49830 amps... What is the minimum heatsink for every amp? I'd like to implement a temperature shutdown for the amps. Which temperature should I choose? Are 65°C ok?
Thanks,
Stammheim
after I couldn't find a solution how to connect both LPUHPs to one transformer, I decided to buy new regulators, as they maybe got damaged by connecting bith rails to one secondary.
As I already wrote the amps do although function, but I wonder about the temperature of the heatsink, which is getting warm after about 5 minutes with only really a little power.
I do also wonder why I have different results for the voltage between A, B and C, as Pwen wrote some posts ago. I hope by changing the regs all problems are gone.
Another question regarding the LME49830 amps... What is the minimum heatsink for every amp? I'd like to implement a temperature shutdown for the amps. Which temperature should I choose? Are 65°C ok?
Thanks,
Stammheim
Hello folks.
I've already burnt through a set of regulators on my PSU, so trying to take things real slow here so please bear with me.
I have a shorting problem and I can't seem to troubleshoot it.
According to the schematics, the two ground planes are supposed to be electrically conducting (as indicated by the heatsink being grounded). Examining the pads for Q3 and Q4 on the PCB, the pins 4 and 5 for both Q3 and Q4 are connected to the ground. But according to the schematics, Pins 4 and 5 for Q3 are grounded, but for Q4 they are insulated and go to +15V. To sum it all up, I have a situation where -15V and ground for Q3 are properly separated, but +15V and ground for Q4 are shorted. Does anyone have any idea of what could be causing this?
I've already burnt through a set of regulators on my PSU, so trying to take things real slow here so please bear with me.
I have a shorting problem and I can't seem to troubleshoot it.
According to the schematics, the two ground planes are supposed to be electrically conducting (as indicated by the heatsink being grounded). Examining the pads for Q3 and Q4 on the PCB, the pins 4 and 5 for both Q3 and Q4 are connected to the ground. But according to the schematics, Pins 4 and 5 for Q3 are grounded, but for Q4 they are insulated and go to +15V. To sum it all up, I have a situation where -15V and ground for Q3 are properly separated, but +15V and ground for Q4 are shorted. Does anyone have any idea of what could be causing this?
Hi Ockam,
It sounds an awful lot like you haven't properly isolated the tab on the Q4 regulator from heatsink it's mounted on.
The heatsinks are both tied to ground for safety, shielding, and thermal reasons, but only the tab of Q3 is at ground potential. The tab of Q4 is at +15V potential, so it needs to be completely electrically isolated from the heatsink it's mounted on.
If you're measuring 0R resistance from Q4 tab to heatsink, then that's probably the issue.
Are you using an electrically isolated thermal pad? Are you using a stepped nylon washer to completely isolate the mounting bolt from the tab?
The only other possibility is a solder short, and we'd need pictures to troubleshoot that.
Cheers,
Owen
It sounds an awful lot like you haven't properly isolated the tab on the Q4 regulator from heatsink it's mounted on.
The heatsinks are both tied to ground for safety, shielding, and thermal reasons, but only the tab of Q3 is at ground potential. The tab of Q4 is at +15V potential, so it needs to be completely electrically isolated from the heatsink it's mounted on.
If you're measuring 0R resistance from Q4 tab to heatsink, then that's probably the issue.
Are you using an electrically isolated thermal pad? Are you using a stepped nylon washer to completely isolate the mounting bolt from the tab?
The only other possibility is a solder short, and we'd need pictures to troubleshoot that.
Cheers,
Owen
Hi Brunk,
The ideal value depends on your source's drive capability. You can toy around with the resistors on the BAL-BAL to get 10K input impedance, and I would suggest using a 10k pot with that.
I personally went out an bought a few of these:
4P 24 Step Rotary Switch Attenuator Volume Control DIY | eBay
I then planned out the resistance values I needed to get a decent log taper and a 10k pot, and bought each value as a 0.1% thin film 0805 resistor. These resistors solder perfectly between each solder lug, resulting in a very short signal path and a very accurate pot.
It was quite a bit of work, but I really like the results. The tactile feel of the switch is nowhere near the quality of a Shallco or an DACT, but it's also nowhere near the price. If you can find a few other people who want kits, you can get the price down to about $37 per balanced stepped attenuator, which is very hard to beat.
Cheers,
Owen
The ideal value depends on your source's drive capability. You can toy around with the resistors on the BAL-BAL to get 10K input impedance, and I would suggest using a 10k pot with that.
I personally went out an bought a few of these:
4P 24 Step Rotary Switch Attenuator Volume Control DIY | eBay
I then planned out the resistance values I needed to get a decent log taper and a 10k pot, and bought each value as a 0.1% thin film 0805 resistor. These resistors solder perfectly between each solder lug, resulting in a very short signal path and a very accurate pot.
It was quite a bit of work, but I really like the results. The tactile feel of the switch is nowhere near the quality of a Shallco or an DACT, but it's also nowhere near the price. If you can find a few other people who want kits, you can get the price down to about $37 per balanced stepped attenuator, which is very hard to beat.
Cheers,
Owen
Daft question time form me...
I have a couple of fully populated 317/337 'The Wire' PSUs but only want to use them for ~ +5V... Would it be best to strip the negative components off the boards, or OK as is with AC connected to positive side only?
Looking forward to hearing more about the 'The Wire' DAC when the time's right ... once they're in the wild I'd like one as part of a balanced headphone amp build. Take it I'll also require the NTD1 board?
Paul
I have a couple of fully populated 317/337 'The Wire' PSUs but only want to use them for ~ +5V... Would it be best to strip the negative components off the boards, or OK as is with AC connected to positive side only?
Looking forward to hearing more about the 'The Wire' DAC when the time's right ... once they're in the wild I'd like one as part of a balanced headphone amp build. Take it I'll also require the NTD1 board?
Paul
Hi Paul,
No need to de-pop those parts. They won't cause any harm staying in the circuit, even with only the positive reg being used.
As for the DAC, I'm at the 95% done stage, but the last 5% is 50% of the work
It's very close to being ready for release, and it will conveniently include an improved version of the NTD1
tiept:
Your name is in the correct spot! I'll send a payment request to you in a few days.
Cheers,
Owen