I am not (yet) well enough familiar with the design, but I do see several power resistors on the BoM...just suggesting it might be a good long term reliability decision to mount on solder side (if that fits) to avoid heating nearby components on the component side. Less heat and gentler thermal cycling is a good thing if you can fit it in.
Last edited:
I would not change where they are mounted. They are fine thermally and probably better ventilated where they are.
Cheers,
Bob
I don't think there is any need for a regulated supply for +/- 15V for the servo. The Zener diodes provide more than enough stability.
Cheers,
Bob
I suggest to build the design as shown and documented. Many have so far and I have had very little feedback which I take as being adequate to get the job done. If improvement in assembly and documentation is required then it will be incorporated into the documents.
R1,2 in the AFE should be at least 1cm raised, since they do get got to touch.
R1,2 in the AFE should be at least 1cm raised, since they do get got to touch.
Last edited:
I will certainly follow instructions to the letter! The docs look great, I particularly appreciate the circuit description (my last adventure was a Luxman).
I ordered BoM’s from Mouser last night, parts arrived 15 min ago. That’s Impressive!
I do have a mouser “exception list” ... parts not deliverable with today’s order...I can post that if there is interest. Most missing items are resistors...and Wurth terminals.
I ordered BoM’s from Mouser last night, parts arrived 15 min ago. That’s Impressive!
I do have a mouser “exception list” ... parts not deliverable with today’s order...I can post that if there is interest. Most missing items are resistors...and Wurth terminals.
This is a problem with a BOM with tons of components, there is always the chance of a back-ordered item. Lots of Nichicon ecaps that I use in other designs are all on back order. It takes time to search for subs.
I know that Vishay 6K8,2W,1% has been on back order for a while.
71-CPF2-F-6.81K-E3 expected 2021-09-20
Imagine as a mfg having to wait this long for special part.
Use 71-CCF2-J-6.8K, buy a few more and select them for a <1% match of the two in series.
Sub 71-CMF60511R00FHEK with 71-CMF60510R00FKEK or 71-CMF60510R00BHR6
Wurth TB's, have to look at getting them at Digi-Key or other disti.
Vishay SFR16S and MBA0204 are basically interchangeable for this design
So are Yaego MFR25F and Vishay MBB0207
Sometimes when I notice a back-order on a BOM, I look at breaking it up, buy some at Mouser, some at Digi-Key. We get the free shipping with >=$100, so try to even it out and not have to pay for shipping.
I know that Vishay 6K8,2W,1% has been on back order for a while.
71-CPF2-F-6.81K-E3 expected 2021-09-20
Imagine as a mfg having to wait this long for special part.
Use 71-CCF2-J-6.8K, buy a few more and select them for a <1% match of the two in series.
Sub 71-CMF60511R00FHEK with 71-CMF60510R00FKEK or 71-CMF60510R00BHR6
Wurth TB's, have to look at getting them at Digi-Key or other disti.
Vishay SFR16S and MBA0204 are basically interchangeable for this design
So are Yaego MFR25F and Vishay MBB0207
Sometimes when I notice a back-order on a BOM, I look at breaking it up, buy some at Mouser, some at Digi-Key. We get the free shipping with >=$100, so try to even it out and not have to pay for shipping.
I have watched Bob Cordell’s 2016 burning amp presentation on 220C a couple of times, and a question arises regarding bulk power to the power amp stage.
When I replace the bulk capacitors on the 220, is there any value gained by splitting the bulk capacitors into two banks separated by a power resistors, similar to what was done to isolate the input and vas stages?
Secondly, if that isn’t worth considering...is the cap value of the Hafler OK...I see many folks increase it (and associated rectifier, and some kind of soft start to cope with the increased capacitance).
My suspicion is that existing cap values are fine left alone, because the 220c boards look after local bulk supply where required.
I would appreciate some comments on this, what decision and why have other builders here implemented?
When I replace the bulk capacitors on the 220, is there any value gained by splitting the bulk capacitors into two banks separated by a power resistors, similar to what was done to isolate the input and vas stages?
Secondly, if that isn’t worth considering...is the cap value of the Hafler OK...I see many folks increase it (and associated rectifier, and some kind of soft start to cope with the increased capacitance).
My suspicion is that existing cap values are fine left alone, because the 220c boards look after local bulk supply where required.
I would appreciate some comments on this, what decision and why have other builders here implemented?
Hi Peter,
Whenever there is a larger reservoir capacitance implemented with 2 or more large capacitors on each rail, I do recommend putting a low-value resistor between them, forming a pi filtering arrangement. This resistor can and should be quite small, perhaps on the order of 0.22 ohms. For example, one might have 10,000 uF followed by 0.22 ohms followed by another 10,000 uF. This arrangement softens the sharp rectification edges that appear on the first capacitor, providing more high frequency attenuation, since the ESR and ESL impedances of the second capacitor is small compared to 0.22 ohms.
Electrolytic capacitor technology has come a long way. It is OK to choose to replace the existing screw-terminal computer electrolytics in the original design with larger ones, like 22,000 uF and not implement the pi arrangement described above. On the other hand, if you implement a small PC board power supply arrangement using snap-in capacitors, then you can use multiple caps per rail and implement the pi arrangement.
Cheers,
Bob
Whenever there is a larger reservoir capacitance implemented with 2 or more large capacitors on each rail, I do recommend putting a low-value resistor between them, forming a pi filtering arrangement. This resistor can and should be quite small, perhaps on the order of 0.22 ohms. For example, one might have 10,000 uF followed by 0.22 ohms followed by another 10,000 uF. This arrangement softens the sharp rectification edges that appear on the first capacitor, providing more high frequency attenuation, since the ESR and ESL impedances of the second capacitor is small compared to 0.22 ohms.
Electrolytic capacitor technology has come a long way. It is OK to choose to replace the existing screw-terminal computer electrolytics in the original design with larger ones, like 22,000 uF and not implement the pi arrangement described above. On the other hand, if you implement a small PC board power supply arrangement using snap-in capacitors, then you can use multiple caps per rail and implement the pi arrangement.
Cheers,
Bob
Hi Peter,
The DH-200/220 chassis is pretty tight so space considerations come into play. You're smart to think about the power supply side of the amp. This is often forgotten and the limiting part of the amp. If you still have the original Sangamo power caps, these are 30+ years old and should be replaced. You can get new power caps that are double the capacitance that are the same physical size that are drop-in replacements. When I replaced these in my DH-500, I increased the capacitance from 20k up to 39k each and immediately noticed improved bass response and dynamics. I consider this one of the biggest bang-for-the-buck that is easy to do. As you point out, increasing the capacitance will increase the in-rush current at turn-on and put a strain on the power switch, transformer and rectifier. I installed a diyAudio soft start board which worked out great. There may not be room in the DH-220 to add this board. An alternative is to use an in-rush limiter thermistor.
I also replaced the bridge rectifier with a IXYS HiperFRED (Mouser part # 747-FBE22-06N1) which reduces transformer ringing along with a transformer CRC snubber. The snubber design is covered in the Quasimodo thread. Thanks to Rinman77 who did the leg work in determining the optimum value for Rx for the Hafler transformers (refer to post #1705 in this thread).
You're in for a real treat with the DH-220C boards. I installed these in my DH-500 and noticed significant improvements in sound quality compared to the Hafler PC19 boards. Thanks to Bob and Rick for their efforts providing these to the DIY audio community!
Bill.
The DH-200/220 chassis is pretty tight so space considerations come into play. You're smart to think about the power supply side of the amp. This is often forgotten and the limiting part of the amp. If you still have the original Sangamo power caps, these are 30+ years old and should be replaced. You can get new power caps that are double the capacitance that are the same physical size that are drop-in replacements. When I replaced these in my DH-500, I increased the capacitance from 20k up to 39k each and immediately noticed improved bass response and dynamics. I consider this one of the biggest bang-for-the-buck that is easy to do. As you point out, increasing the capacitance will increase the in-rush current at turn-on and put a strain on the power switch, transformer and rectifier. I installed a diyAudio soft start board which worked out great. There may not be room in the DH-220 to add this board. An alternative is to use an in-rush limiter thermistor.
I also replaced the bridge rectifier with a IXYS HiperFRED (Mouser part # 747-FBE22-06N1) which reduces transformer ringing along with a transformer CRC snubber. The snubber design is covered in the Quasimodo thread. Thanks to Rinman77 who did the leg work in determining the optimum value for Rx for the Hafler transformers (refer to post #1705 in this thread).
You're in for a real treat with the DH-220C boards. I installed these in my DH-500 and noticed significant improvements in sound quality compared to the Hafler PC19 boards. Thanks to Bob and Rick for their efforts providing these to the DIY audio community!
Bill.
@HaflerDh500Fan
I'm looking to recap my DH-200 too and was considering the Musical Concepts boards using "fancy" Mundorf AG (2 pin 15,000uF 100V on sale) or AG+ (4 pin 10,000uF 80V) caps complete with the IXYS bridge.
I wanted to use the "doubled up" PS-220 design but my amp was partially "Pooged" when it was built and the polypropylene input caps I installed actually take up a lot of space so I am constrained to the simpler PS-100.
Any thoughts on the capacitor options?
Would the snubber you mention work on the stock DH-200 transformer? I tried searching for the thread and the post but failed [sigh]. Do you have a link to this to help me out?
I'm looking to recap my DH-200 too and was considering the Musical Concepts boards using "fancy" Mundorf AG (2 pin 15,000uF 100V on sale) or AG+ (4 pin 10,000uF 80V) caps complete with the IXYS bridge.
I wanted to use the "doubled up" PS-220 design but my amp was partially "Pooged" when it was built and the polypropylene input caps I installed actually take up a lot of space so I am constrained to the simpler PS-100.
Any thoughts on the capacitor options?
Would the snubber you mention work on the stock DH-200 transformer? I tried searching for the thread and the post but failed [sigh]. Do you have a link to this to help me out?
Last edited:
Hi Peter,
I have looked at many retrofits in my DH-220C journey. There are many options available and in use. Some just change the screw term ecaps, add film ecpas, bleed resistors, clean up the wiring. Others gut the old PS and put in a new toroid transformer ... Some add little expense, some great expense. In some cases it really depends on what you want out of the amplifier, the levels, the loads that it has to drive.
My DH-200 test unit is stock, it will need a PS retrofit at some point. I guess I should at least measure those old ecaps for fit. It has been on since I installed the upgrade. Long term reliability testing, makes a nice hand warmer as well. I think my LCR can measure those ecaps, time to check
I have been pondering the PS upgrade options as well. Bob uses an Asian supplied rectifier/filter, which is a good economical option.
I myself would rather design a new PS pcb that incorporates what Bill has described above. Possibly could also include a speaker turn-on/off relay or mosfet, DC offset detection. It is a tight fit, but if done correctly, it should work out okay. It is the mechanical details that need attention. If one uses the existing holes in the chassis, might not need to do any drilling etc.
Throw out your ideas, for a new PS pcb design for the DH-200/220 chassis. I could do one, since I was planning on one anyways. One problem with this approach is everyone has their own ideas, getting agreement, but we can give it a shot.
I have the building blocks in place, here is one incarnation that we are doing up for a DIYAudio 3Ux300 chassis.
I myself do would rather use the STmicro FERD parts vs the IXYS parts. Compare the specs and see if you agree
Rick
I have looked at many retrofits in my DH-220C journey. There are many options available and in use. Some just change the screw term ecaps, add film ecpas, bleed resistors, clean up the wiring. Others gut the old PS and put in a new toroid transformer ... Some add little expense, some great expense. In some cases it really depends on what you want out of the amplifier, the levels, the loads that it has to drive.
My DH-200 test unit is stock, it will need a PS retrofit at some point. I guess I should at least measure those old ecaps for fit. It has been on since I installed the upgrade. Long term reliability testing, makes a nice hand warmer as well. I think my LCR can measure those ecaps, time to check
I have been pondering the PS upgrade options as well. Bob uses an Asian supplied rectifier/filter, which is a good economical option.
I myself would rather design a new PS pcb that incorporates what Bill has described above. Possibly could also include a speaker turn-on/off relay or mosfet, DC offset detection. It is a tight fit, but if done correctly, it should work out okay. It is the mechanical details that need attention. If one uses the existing holes in the chassis, might not need to do any drilling etc.
Throw out your ideas, for a new PS pcb design for the DH-200/220 chassis. I could do one, since I was planning on one anyways. One problem with this approach is everyone has their own ideas, getting agreement, but we can give it a shot.
I have the building blocks in place, here is one incarnation that we are doing up for a DIYAudio 3Ux300 chassis.
I myself do would rather use the STmicro FERD parts vs the IXYS parts. Compare the specs and see if you agree
Rick
Last edited:
Is there any additional benefit to using an inductor instead of a resistor? I'm thinking of one of a 5 or 10A rated uH inductor with resistance on the same order as .22 ohms.
@mhenschel-
The Musical Concept PS boards are nice. Although I'm surprised that two PS-220 boards could fit in the DH-200 chassis. The Mundorf AG caps are quite good although they are pricey. My advice would be to go with the drop in replacement caps first. This is relatively inexpensive mod and easy to do.
Yes, the CRC snubber will work on the DH-200 transformer. In fact, post #1705 of this thread (page 171) has the info you need for the snubber design. Also, the link to the Quasimodo thread is in this same post.
The Musical Concept PS boards are nice. Although I'm surprised that two PS-220 boards could fit in the DH-200 chassis. The Mundorf AG caps are quite good although they are pricey. My advice would be to go with the drop in replacement caps first. This is relatively inexpensive mod and easy to do.
Yes, the CRC snubber will work on the DH-200 transformer. In fact, post #1705 of this thread (page 171) has the info you need for the snubber design. Also, the link to the Quasimodo thread is in this same post.
@Halauhula-
Yes, a CLC is better than a CRC for reducing PS ripple voltage. However, good luck trying to fit the inductors in the Hafler chassis. The 10A rated inductors are rather large. Ozark HiFi talks about testing CLC in post #1964. It would be interesting to hear the status of this. I suspect the CLC would need to be in a separate chassis.
Yes, a CLC is better than a CRC for reducing PS ripple voltage. However, good luck trying to fit the inductors in the Hafler chassis. The 10A rated inductors are rather large. Ozark HiFi talks about testing CLC in post #1964. It would be interesting to hear the status of this. I suspect the CLC would need to be in a separate chassis.