Will the Modulus-86 be unstable if the Burson V6 is also used for U3?
Now using it in U2 and I like this move.
Ad.
There isn't enough information in the Burson data sheet for me to answer that question. Specifically I will need the AVOL curve (open loop gain/phase curve).
Note that many discrete opamps have a second order (-40 dB/dec) rolloff with an internal compensation making them unity gain stable. Such an opamp can provide good performance at unity gain, but will have significant ringing on the transient response (assuming they're stable at all) when used away from unity gain. I don't know if the Burson falls in this category as they don't show that data in the data sheet.
Tom
Note that many discrete opamps have a second order (-40 dB/dec) rolloff with an internal compensation making them unity gain stable. Such an opamp can provide good performance at unity gain, but will have significant ringing on the transient response (assuming they're stable at all) when used away from unity gain. I don't know if the Burson falls in this category as they don't show that data in the data sheet.
Tom
Hey @tomchr!
After building that "Gundam" HP-1 and an HP-2 for a buddy, I'm back again looking at some more efficient amps (Class AB or D) to run during the summer months. Thought of your projects after finding how great those headlamps are.
Question that I've tried searching for but couldn't find a clear answer... I have a pair of 250VA Triad medical-grade toroids that are 24VAC secondaries. Could I stuff these and modulus-86's into a 4U monoblock for a dual mono design (stacked PSUs)? Or would 24VAC be too high? Seems like 250VA @ 24VAC (x2) may work nicely with with Modulus-286?
Would be awesome to repurpose these nice xformers for a build instead of having to sell them on to someone else.
After building that "Gundam" HP-1 and an HP-2 for a buddy, I'm back again looking at some more efficient amps (Class AB or D) to run during the summer months. Thought of your projects after finding how great those headlamps are.
Question that I've tried searching for but couldn't find a clear answer... I have a pair of 250VA Triad medical-grade toroids that are 24VAC secondaries. Could I stuff these and modulus-86's into a 4U monoblock for a dual mono design (stacked PSUs)? Or would 24VAC be too high? Seems like 250VA @ 24VAC (x2) may work nicely with with Modulus-286?
Would be awesome to repurpose these nice xformers for a build instead of having to sell them on to someone else.
Last edited:
2x24 VAC would be good for an 8 Ω optimized Modulus-86. You'll get around ±33 V rectified so with a 4 Ω load you'll see the LM3886 trigger its SPiKe protection pretty early, especially on bass thumps. ±28-30 V is about optimal for the LM3886.
2x24 VAC would be perfect for a Modulus-286 amp, though. You could power a stereo amp with that or use it in a mono block.
I'm currently out of the Modulus-286 but can have more built in a matter of days. I complete the final mechanical assembly on those so it's just a matter for me to clear my lab bench and sit down and turn a screwdriver for a few hours.
Tom
2x24 VAC would be perfect for a Modulus-286 amp, though. You could power a stereo amp with that or use it in a mono block.
I'm currently out of the Modulus-286 but can have more built in a matter of days. I complete the final mechanical assembly on those so it's just a matter for me to clear my lab bench and sit down and turn a screwdriver for a few hours.
Tom
Awesome. Is this also the same case with the Modulus-686 (safe-n-sane)?
I have an affinity with higher sensitivity speakers. The lowest sensitivity I'll likely end up using is 93dB/W, 4Ω. But, I could very easily end up with horns someday too. So, I don't need gobs of power, but I do appreciate lower noise floors and over sound quality (loaded term, I know.)
Given this, would you recommend the 286 or 686 (safe-n-sane)?
Also, could I get away with this LT4320 PSU? My rectifiers (MOSFETs) can definitely handle the current. My CRC is 33mF (35WVDC), 0.1Ω(10W), 33mF. Two boards, so two rails per channel. Already have 'em built so it'd be great to put them to action, especially since your Modulus amps seem to have high PSRR. Thoughts?
I'm definitely interested in a few of your soft start boards, and speaker protection. Dang, this hobby can get expensive lol.
I am not understand how the Burson V6 is integrated on a PCB but photos on the burson site show the V6, mounted on PCBs, not connected to any heatsink. Does replacing the lm3886 with a burson V6 eliminate the need for a heatsink?
You can go with a 2U chassis. I offer one: https://neurochrome.com/collections/connectors-parts/products/modulus-chassis
It's currently out of stock, but I can order more.
If you don't need to blow the windows out go for the Modulus-286. Or Modulus-86 with a 2x22 VAC transformer and Power-86.
Sure. I'd replace the R with a wire link though.
I have plenty of those.
He's referring to the part designators on Rev. 3.0 of the Modulus-86. U2 is acting as an instrumentation amp front-end. U3 is the error correction opamp.
The Burson is an opamp. Not a replacement for the LM3886.
Tom
Very confusing. The bill of materials lists U2 as an LM3886 and the A-3: mechanical dimensions seems to show U2 as a device with many connections. Obviously way beyond my understanding.
Sorry. It was @mkrawcz who mentioned the Burson and Rev. 3: https://www.diyaudio.com/community/threads/modulus-86-build-thread.267802/post-6906588
Tow
Tow
How tall is the chassis? One option could be to mount the MOD86 boards vertically and secure them with long standoffs to the rear (or front) panel.
Of the options you present I prefer Option 2, though I would consider mounting the rectifier on the bottom panel between the transformers. Then run the mains wiring straight up the middle of the chassis. Here's my rationale:
Option 1 doesn't allow you to get the wires into the terminal blocks. No bueno!
Option 3 runs the wires for the rectifier (nasty pulse currents) right over the input circuit for the left channel. That'll give hum.
Option 2 does have the (EI core!) transformer right by the input circuit on the right channel, but given the choice between having some stray electromagnetic field from the transformer engage with the input circuit (Option 2) and having the EM field from the rectifier currents do so (Option 3) I think Option 2 has the lower chance of inducing hum. That said, the right channel in Option 2 will measure worse that the left because of the coupling from the transformer. Sorry. Physics... 🙂
Placing the rectifier between the transformers in Option 2 and rotating the PSU board such that the wiring to the rectifier is as short as possible will limit the loop area in the rectifier harness, thereby minimizing its coupling into the rest of the circuitry. Keep the wires short and bundle them together with zip ties.
Tom
Of the options you present I prefer Option 2, though I would consider mounting the rectifier on the bottom panel between the transformers. Then run the mains wiring straight up the middle of the chassis. Here's my rationale:
Option 1 doesn't allow you to get the wires into the terminal blocks. No bueno!
Option 3 runs the wires for the rectifier (nasty pulse currents) right over the input circuit for the left channel. That'll give hum.
Option 2 does have the (EI core!) transformer right by the input circuit on the right channel, but given the choice between having some stray electromagnetic field from the transformer engage with the input circuit (Option 2) and having the EM field from the rectifier currents do so (Option 3) I think Option 2 has the lower chance of inducing hum. That said, the right channel in Option 2 will measure worse that the left because of the coupling from the transformer. Sorry. Physics... 🙂
Placing the rectifier between the transformers in Option 2 and rotating the PSU board such that the wiring to the rectifier is as short as possible will limit the loop area in the rectifier harness, thereby minimizing its coupling into the rest of the circuitry. Keep the wires short and bundle them together with zip ties.
Tom
The photos show the leads from the IEC to the ISS running over the right channel parallel86. Could that be a problem? Should I run the wires inside some sheilding?
I would snip the ground wire from the ISS to the chassis. The ISS does not need to be grounded and cutting that wire would eliminate any leakage currents from the mains filter in the SMPS on the ISS.
You could argue that the mains wiring should run down the middle between the heat sinks for similar reasons as in my advice just above. On the other hand: Do you have audible hum in the speakers when you're using this amp? If not, don't mess with it. Ain't broken. Don't fix.
Tom
You could argue that the mains wiring should run down the middle between the heat sinks for similar reasons as in my advice just above. On the other hand: Do you have audible hum in the speakers when you're using this amp? If not, don't mess with it. Ain't broken. Don't fix.
Tom