Canada post still hasn’t even acknowledged picking up the delivery, let alone delivering the boards. Sigh. “Shipping label created” is where it sits.
I dissembled my 220 yesterday, it is in very good shape, clean inside. Lots to do to get chassis ready while I await boards, and I’ll order the few missing bits I need today.
Some of the internal wiring has crumbling insulation, but aside from that it’s great. The Sangamo bulk caps are high on esr, so I will be replacing. Interesting they actually printed “misapplication dangerous” on those caps.
Box itself is in good shape but heat sinks are scratched up. Might look to getting them resurfaced.
I dissembled my 220 yesterday, it is in very good shape, clean inside. Lots to do to get chassis ready while I await boards, and I’ll order the few missing bits I need today.
Some of the internal wiring has crumbling insulation, but aside from that it’s great. The Sangamo bulk caps are high on esr, so I will be replacing. Interesting they actually printed “misapplication dangerous” on those caps.
Box itself is in good shape but heat sinks are scratched up. Might look to getting them resurfaced.
To add to the discussion of SMPS, there's a diyAudio thread entitled "switch mode power supply for good amplifiers" in the Forums>Amplifiers>Power Supplies section. In it, there's reference to Hypex that offers SMPS for amps. Looking at their web site, they do offer a 64V SMPS that could be considered for the DH220.
I also noticed in a different thread ... "Modulus-686...", there's some posts about using SMPS. Post 837 shows using a Mean Well SMPS for a Neurochrome Modulus-686 amp. Although this amp has much lower rail voltages than the DH220 (27v vs 64v). It appears the Mean Well SMPSs only go up to 48v. Anyway, it's another example that SMPS can be used in high quality power amps.
I also noticed in a different thread ... "Modulus-686...", there's some posts about using SMPS. Post 837 shows using a Mean Well SMPS for a Neurochrome Modulus-686 amp. Although this amp has much lower rail voltages than the DH220 (27v vs 64v). It appears the Mean Well SMPSs only go up to 48v. Anyway, it's another example that SMPS can be used in high quality power amps.
I will just complete my effort on my own.
I don’t like the tone of this thread, so I will watch for content that’s useful and fill my needs other ways.
Thanks for help so far. Hopefully I haven’t just wasted mine and everyone else’s time. Apologies for any unintended offence, but I am incompatible with this.
I remain excited about the project, and the chassis challenges to come. Many thanks for the release of this engineering effort.
Peter.
Merry Christmas and stay healthy.
I don’t like the tone of this thread, so I will watch for content that’s useful and fill my needs other ways.
Thanks for help so far. Hopefully I haven’t just wasted mine and everyone else’s time. Apologies for any unintended offence, but I am incompatible with this.
I remain excited about the project, and the chassis challenges to come. Many thanks for the release of this engineering effort.
Peter.
Merry Christmas and stay healthy.
SMPS Success!
I removed the stock transformer and power supply, installed a Cresnet SMPS-600, and made all the logical connections. I kept the stock circuit for the mains fuse and power switch. Also took the opportunity with the old power supply out to enlarge the RCA jack openings and install better jacks.
The SMPS measured 67VDC unloaded. After I made sure everything was operational and even played test music through it for a bit, I buttoned up the top and have been playing more music with the test speakers, which admittedly sound pretty darn good (tang band full range drivers in minimus 7 cabinets).
After about 45 minutes of operating well without any problems, the heatsinks are suitably warm, just as they would have been with the stock power supply. Maybe a couple degrees warmer (around 122F/50C).
It sounds just as good as before, possibly a tad bit more refined, with more discernible mid and high detail. But that’s all based on memory, fatigue, and elation at a successful tinkering.
It’s also dead silent, which is why I wanted to try this. The stock transformer had developed a mechanical hum that really annoyed me.
So let this be proof positive that a DH-220C can work well with a SMPS!
I am, as always, grateful for Bob’s and Rick’s generous offering of the DH-220C circuit, boards, and support for the diyAudio community regarding this amplifier. Plus the ongoing advise, experimentation, and ingenuity of those contributing to this great thread. Long live the lat-FET Haflers!
I removed the stock transformer and power supply, installed a Cresnet SMPS-600, and made all the logical connections. I kept the stock circuit for the mains fuse and power switch. Also took the opportunity with the old power supply out to enlarge the RCA jack openings and install better jacks.
The SMPS measured 67VDC unloaded. After I made sure everything was operational and even played test music through it for a bit, I buttoned up the top and have been playing more music with the test speakers, which admittedly sound pretty darn good (tang band full range drivers in minimus 7 cabinets).
After about 45 minutes of operating well without any problems, the heatsinks are suitably warm, just as they would have been with the stock power supply. Maybe a couple degrees warmer (around 122F/50C).
It sounds just as good as before, possibly a tad bit more refined, with more discernible mid and high detail. But that’s all based on memory, fatigue, and elation at a successful tinkering.
It’s also dead silent, which is why I wanted to try this. The stock transformer had developed a mechanical hum that really annoyed me.
So let this be proof positive that a DH-220C can work well with a SMPS!
I am, as always, grateful for Bob’s and Rick’s generous offering of the DH-220C circuit, boards, and support for the diyAudio community regarding this amplifier. Plus the ongoing advise, experimentation, and ingenuity of those contributing to this great thread. Long live the lat-FET Haflers!
Attachments
Congratulations! This is a great accomplishment! Thank you for sharing it with us.
How much did the Cresnet SMPS-600 cost?
You mentioned that the SMPS measured 67 V rails no loaded. Was this truly unloaded, or under conditions of being loaded with the quiescent bias current of both channels. I'm curious about the regulation of the SMPS-600.
Cheers,
Bob
How much did the Cresnet SMPS-600 cost?
You mentioned that the SMPS measured 67 V rails no loaded. Was this truly unloaded, or under conditions of being loaded with the quiescent bias current of both channels. I'm curious about the regulation of the SMPS-600.
Cheers,
Bob
Thanks, Bob. I got this SMPS from a fellow forum member and paid ~90 USD for two of them. According to Cresnet, this is an older model. If I recall correctly, it is also unregulated, which I thought would in theory work well with the Hafler since the original PS was unregulated. The 67VDC output was measured completely unloaded. I should measure the output with everything connected...
In looking at the specs on this SMPS I saw something very unexpected, but which should have been obvious to me. I'm referring to the scope pictures that show significant 100-Hz ripple on the output of the SMPE under load. My initial knee-jerk reaction was - what? SMPS should have low ripple and it should be at a high frequency. Then I remembered that this SMPS is unregulated, unlike many we are used to. Here is what is apparently going on. The SMPS first rectifies the line voltage and stores the resulting DC on a reservoir capacitor on the intermediate DC bus ahead of the guts of the SMPS (on the primary side). I think we are looking at the ripple on the intermediate bus coming through because the SMPS is unregulated. If the intermediate bus reservoir capacitor were doubled, I bet the ripple on the output side would be halved.
Cheers,
Bob
Cheers,
Bob
SMPS Success!
I removed the stock transformer and power supply, installed a Cresnet SMPS-600, and made all the logical connections. I kept the stock circuit for the mains fuse and power switch. Also took the opportunity with the old power supply out to enlarge the RCA jack openings and install better jacks.
The SMPS measured 67VDC unloaded. After I made sure everything was operational and even played test music through it for a bit, I buttoned up the top and have been playing more music with the test speakers, which admittedly sound pretty darn good (tang band full range drivers in minimus 7 cabinets).
After about 45 minutes of operating well without any problems, the heatsinks are suitably warm, just as they would have been with the stock power supply. Maybe a couple degrees warmer (around 122F/50C).
It sounds just as good as before, possibly a tad bit more refined, with more discernible mid and high detail. But that’s all based on memory, fatigue, and elation at a successful tinkering.
It’s also dead silent, which is why I wanted to try this. The stock transformer had developed a mechanical hum that really annoyed me.
So let this be proof positive that a DH-220C can work well with a SMPS!
I am, as always, grateful for Bob’s and Rick’s generous offering of the DH-220C circuit, boards, and support for the diyAudio community regarding this amplifier. Plus the ongoing advise, experimentation, and ingenuity of those contributing to this great thread. Long live the lat-FET Haflers!
Congratulations you have made a great job in modifying the DH-220 amp! This was an ambitious project.
Fab
Exactly. I think the best outcome noise wise could be achieved from SMPS's with PFC, because these have the intermediate voltage regulated also.
Anyway, ripple in unregulated SMPS's will occur with heavy loads only, and as this occurs at higher listening levels, it will be swamped by the signal - just with unregulated conventional PSU's.
Best regards!
Last edited:
Thanks for your comments. It was a bit of a gamble on my part, but I knew from other threads that these MicroAudio SMPS are designed for audio amplifiers and have had positive reviews in the forum. At the very least, I knew I needed to update the stock power supply and if the SMPS didn't pan out, I could go back to a linear supply with all new parts. My gamble has paid off so far!
Yes, PFC would probably help with the noise emittion from the SMPS as well.
Perhaps Van can tell us the value of the intermediate bus capacitor and how accessible it might be for paralleling an external one to it. At the expense of larger inrush current, it might be preferable to augment that one so that very large capacitance on the load side does not interfere with start-up behavior of the SMPS. Just speculating ... As an example, a fair question is "how does the SMPS behave if we increase the secondary capacitance on each rail to 20,000 or 30,000 uF?
Cheers,
Bob
Attached are pics of the second identical unit I have. Further questions on its design, limitations, etc. should be directed to Cresnet. Audio SMPS units for sale
I was informed previously that this particular model should not have capacitance after the output exceeding 4700uF.
I was informed previously that this particular model should not have capacitance after the output exceeding 4700uF.
Attachments
Last edited:
SMPS600 cannot tolerate more than 4700uF as TOTAL per rail of extra capacitance.
However, the SMPS630-G can take up to 30000uF per rail, and the SMPS500-CLA for class-A amplifiers takes up 75000uF per rail.
Adding extra capacitors at the output will decrease the ripple dramatically.
Hope that helps
Attachments
Last edited:
Thanks for this information. How big is the intermediate bus capacitor(s)? Can you describe the line side rectification circuitry? Does this circuitry provide power factor correction?
Cheers,
Bob