Bob: I may have missed an earlier post, but I'm curious about the BC-1 PCBs. Are they available yet? I so like the modular design idea. With them, I can upgrade the INPUT-VAS-predriver board of an old 250W Amp I have that conformed to a bad idea in the mid 80s - low feedback, low loop gain. Urei 6300. Otherwise it has a nice power supply, thermal design, output transistor arrangement. Pretty poor distortion performance however.
Bob’s Christmas present from me for 2021, Santa’s work shop runs quietly all year. Well sort of like that as shown, he gets to assembly his own from the blanks Santa gave him. I actually powered one channel up today, it holds bias and dc offset is <10mV, lots of work left to do but I think we are getting closer, bit by bit, winter is set in, iron is hot, happy new year too all.
On the Hafler front, I finally fixed my old fried power switch, went one step further and designed a soft start to save the new switch contacts from frying again.
On the Hafler front, I finally fixed my old fried power switch, went one step further and designed a soft start to save the new switch contacts from frying again.
Attachments
When can I get a set of boards?
Is the comparator/protection/relay pcb also available?
I just purchased the 220C set, but am also interested in the BC-1 set. They look great.
One other question: What's the shortest dimension of the Input/VAS pcb? i.e. Height in the last picture.
BC-1 is still in the works, release some time in I hope winter/spring 2022. It is being designed for the 3Ux300mm chassis. There is a protection & PS pcbs, just like in the book. Nothing released until Bob gives it his blessing, just like DH-220C 
Just wanted to show you we are moving along, slowly, since you asked.
Good luck with DH-220C, I have mine running since the day I fired it up. Glad I fixed the AC switch issue, now I have to change out those old RCA jacks. I will try these Neutrik NYS367-2 & NYS367-9. The cheap ones from Asia just don't cut it for me.
Just wanted to show you we are moving along, slowly, since you asked. Good luck with DH-220C, I have mine running since the day I fired it up. Glad I fixed the AC switch issue, now I have to change out those old RCA jacks. I will try these Neutrik NYS367-2 & NYS367-9. The cheap ones from Asia just don't cut it for me.
Too bad. I'm planning to cram the BC-1 into a 2U chassis from a dead Urei 6250. I guess the wider input/VAS board needn't be on the side heat sinks. Thanks for responding so quickly. Happy Holidays!BC-1 is still in the works, release some time in I hope winter/spring 2022. It is being designed for the 3Ux300mm chassis. There is a protection & PS pcbs, just like in the book. Nothing released until Bob gives it his blessing, just like DH-220C
Good luck with DH-220C, I have mine running since the day I fired it up. Glad I fixed the AC switch issue, now I have to change out those old RCA jacks. I will try these Neutrik NYS367-2 & NYS367-9. The cheap ones from Asia just don't cut it for me.
I was gifted a Hafler DH-200 from my brother this Xmas. He also gave me a Hafler Iris Pre-amp that required some repair (which I have done so).
I think I would like to attempt Bob's DH-220C upgrade. Question - Is there any special consideration when upgrading the DH-200 vs. DH-220?
Also, I am looking for a Hafler Iris remote control if anybody has one laying around
I think I would like to attempt Bob's DH-220C upgrade. Question - Is there any special consideration when upgrading the DH-200 vs. DH-220?
Also, I am looking for a Hafler Iris remote control if anybody has one laying around
Nice gift!!
For sure, it can be done in a DH-200, since I used a couple of DH-200's for my proto builds, Bob has DH-220 and P-230 chassis, Others put them in DH-500, still waiting for a P500/505 mod to show up.
Photo above on the right is the new soft-start pcb/ckt installed in the DH-200 chassis. I also changed the power switch, since it was fried.
I am still moding my DH-200, thinking about what I'll do for the ecap upgrade, it will be some 50mm CDE snap-in's, since I have them and they fit the existing clamps.
I cant help you with the remote other than show you how to make your own, just need to figure out the protocol and codes the original sent out, the fun part There might be other options on the remote, but that is MCU and programming stuff which has a learning curve associated with it.
Good Luck
Rick
For sure, it can be done in a DH-200, since I used a couple of DH-200's for my proto builds, Bob has DH-220 and P-230 chassis, Others put them in DH-500, still waiting for a P500/505 mod to show up.
Photo above on the right is the new soft-start pcb/ckt installed in the DH-200 chassis. I also changed the power switch, since it was fried.
I am still moding my DH-200, thinking about what I'll do for the ecap upgrade, it will be some 50mm CDE snap-in's, since I have them and they fit the existing clamps.
I cant help you with the remote other than show you how to make your own, just need to figure out the protocol and codes the original sent out, the fun part
There might be other options on the remote, but that is MCU and programming stuff which has a learning curve associated with it.Good Luck
Rick
So I guess the FETS on the 200 are not much different than the 220?
Yep - remote is pretty rare. Being an embedded guy I was considering interfacing an ESP32 board to the IRIS. Then I could connect it to wifi and control it with my cell phone. This is something I have done before for my weather station. Assuming I could reverse engineer the control board on the IRIS. Possibly a subject for another thread.
Yep - remote is pretty rare. Being an embedded guy I was considering interfacing an ESP32 board to the IRIS. Then I could connect it to wifi and control it with my cell phone. This is something I have done before for my weather station. Assuming I could reverse engineer the control board on the IRIS. Possibly a subject for another thread.
I have the schematic for the Iris. It's not the best schematic and it does not include what is on the IR remote daughter board but there is enough that I can discern that my ESP32 plan looks doable. I'll need to convert from 3v logic to "0V low 14-16V high U.O.S"
Selecting from the various line inputs looks easy...a few mux lines.
I suspect on the remote daughter board is a serial data register that gets loaded via the IR link. But It is still not clear how the remote volume and balance work. I know that there are another set of CLM 8500 devices on the daughter board so there may be associated D to A converters as well.
Fortunately the remote daughter board is not very large and does not look too complicated.
I'll start a new thread regarding this project. I am using the pre-amp presently so I'll have to pull it out of the deck to take a look inside again.
Selecting from the various line inputs looks easy...a few mux lines.
I suspect on the remote daughter board is a serial data register that gets loaded via the IR link. But It is still not clear how the remote volume and balance work. I know that there are another set of CLM 8500 devices on the daughter board so there may be associated D to A converters as well.
Fortunately the remote daughter board is not very large and does not look too complicated.
I'll start a new thread regarding this project. I am using the pre-amp presently so I'll have to pull it out of the deck to take a look inside again.
Getting back to the DH-220C upgrade...
It would have been nice if you and Bob had compared the original DH-200/220 in some of the performance plots that were in the AudioExpress articles.
I was just looking at the specs for the DH-200 and they aren't that bad to begin with. But these specs are different than what has been demonstrated for the DH-220C Some areas I can see are much better, like the dampening factor, better slew rate, and running the FETS at a higher bias to extend the Class A region.
So, Rick Let me ask you .. What performance measure of the DH-220C really stands out as a big improvement to you when compared to the original?
I think the other more pertinent question is if the Halfer IRIS pre-amp is up to snuff to feed the DH-220C upgrade?
It would have been nice if you and Bob had compared the original DH-200/220 in some of the performance plots that were in the AudioExpress articles.
I was just looking at the specs for the DH-200 and they aren't that bad to begin with. But these specs are different than what has been demonstrated for the DH-220C Some areas I can see are much better, like the dampening factor, better slew rate, and running the FETS at a higher bias to extend the Class A region.
So, Rick Let me ask you .. What performance measure of the DH-220C really stands out as a big improvement to you when compared to the original?
I think the other more pertinent question is if the Halfer IRIS pre-amp is up to snuff to feed the DH-220C upgrade?
Last edited:
OK... Here is my concern...
In the video at Burning Amp at around the 1:11 mark.
Bob presents a graph of THD as he measured on the Hafler prior to making his modifications.
I believe he indicates for that plot that the THD data is measured at 1khz,
I find that this contradicts what the Hafler manual states for the DH-200 in the specs, Namely...
"Typical THD at 100 watts into 8 ohms.
1khz 0.0015%
10khz 0.005%
20khz 0.012%"
What am I missing here? Bob's plot indicates 0.3% at 100 watts @ 1khz
Of course I should mention that even the Hafler specifications are a little ambiguous because they also state...
"Power Rating: Less than 0.02% total harmonic distortion at any power level up to 100 watts continuous average power
per channel into 8 ohms at any frequency between 20hz to 20khz with both channels being driven.
But this is still much less than the 0.3% THD that Bob indicates.
Not doubting all the improvements in Bob's new design, as it is all clearly explained in his presentation.
I just don't understand the discrepancy above.
In the video at Burning Amp at around the 1:11 mark.
I believe he indicates for that plot that the THD data is measured at 1khz,
I find that this contradicts what the Hafler manual states for the DH-200 in the specs, Namely...
"Typical THD at 100 watts into 8 ohms.
1khz 0.0015%
10khz 0.005%
20khz 0.012%"
What am I missing here? Bob's plot indicates 0.3% at 100 watts @ 1khz
Of course I should mention that even the Hafler specifications are a little ambiguous because they also state...
"Power Rating: Less than 0.02% total harmonic distortion at any power level up to 100 watts continuous average power
per channel into 8 ohms at any frequency between 20hz to 20khz with both channels being driven.
But this is still much less than the 0.3% THD that Bob indicates.
Not doubting all the improvements in Bob's new design, as it is all clearly explained in his presentation.
I just don't understand the discrepancy above.
Last edited:
D. Hedin , be grateful for Bob’s new design. He’s also busy on commercial projects and his efforts and Ricks are greatly appreciated. If you want a fully documented comparison of what ever details interest you your welcome to test them at your leisure. If you looked at the Hafler schematic there is a global feedback loop, but who knows what exact methodology was used for testing, it was over 40 years ago. The original DH 200 was an Erno Borbely design.
ticknpop, don't get me wrong. I am not doubting Bob's and Ricks efforts. I am certain that new design design is a vast improvement.
I am just trying to understand the discrepancy.
Going from .0015% THD @ 1khz to 0.3% THD @ 1khz is a big difference. That is a factor of 200
Are the old specs a lie?
Is the majority of the difference attributed to noise?
I am just trying to understand the discrepancy.
Going from .0015% THD @ 1khz to 0.3% THD @ 1khz is a big difference. That is a factor of 200
Are the old specs a lie?
Is the majority of the difference attributed to noise?
Looking at the DH200 manual that I downloaded from the web, the Specifications (p. 16) say: "Typical THD at 100 watts into 8 ohms."
Not sure what "typical" in this context but this may suggest that Hafler Corp. was using a non-standard approach to characterize the distortion. I say "may" because I am not an expert in this area of testing.
In any event, many audiophiles and experts in this area argue that THD is not the best measure of the audible quality of audio equipment. They suggest that IM distortion, dynamic capability, for example, may be better or at least a supplemental figures of merit for audio equipment.
Not sure what "typical" in this context but this may suggest that Hafler Corp. was using a non-standard approach to characterize the distortion. I say "may" because I am not an expert in this area of testing.
In any event, many audiophiles and experts in this area argue that THD is not the best measure of the audible quality of audio equipment. They suggest that IM distortion, dynamic capability, for example, may be better or at least a supplemental figures of merit for audio equipment.
Halahula - Agreed THD is not everything. It is probably best to consider the pre-amp, amp, and speakers as a total system. The high slew rate and better damping factor of Bob's new design are reason enough for the upgrade from my perspective. I also believe it is a "good thing" to have the main amp perform close to the performance of the pre-amp. If it exceeds, that is still OK, room to upgrade the pre-amp in the future!
But I am still wondering about the THD discrepancy. It is just so large. I am sure it has to do with how things are measured, but I don't understand what that difference is.
Here's another one for you. For my IRIS pre-amp Hafler indicates the following THD+N:
Moving Magnet: .002%
Moving Coil: .009%
But every analog input gets routed through a CMOS MUX (CD4051). The best performing (THD) CD4051 I can find is the old National part.
Which they indicate is 0.04% @ 1khz into a 10kohm load. (BTW, TI offers the contemporary version of the CD4051 and it's THD performance is much worse)
So how does Hafler get to the .002-.009 range? Is it the case that the output of the CD4051 driving into a higher impedance of the Hafler JFET amp makes the THD of the CD4051 improve? Not so sure.
But I am still wondering about the THD discrepancy. It is just so large. I am sure it has to do with how things are measured, but I don't understand what that difference is.
Here's another one for you. For my IRIS pre-amp Hafler indicates the following THD+N:
Moving Magnet: .002%
Moving Coil: .009%
But every analog input gets routed through a CMOS MUX (CD4051). The best performing (THD) CD4051 I can find is the old National part.
Which they indicate is 0.04% @ 1khz into a 10kohm load. (BTW, TI offers the contemporary version of the CD4051 and it's THD performance is much worse)
So how does Hafler get to the .002-.009 range? Is it the case that the output of the CD4051 driving into a higher impedance of the Hafler JFET amp makes the THD of the CD4051 improve? Not so sure.
Seems I have found an answer to my question above...Yes you can improve the THD of an analog CMOS switch. Great read, BTW.
https://www.ti.com/lit/an/slyt612/slyt612.pdf
..although it is hard for me to tell with the crude IRIS schematic I have if they are including the CMOS switch in the amp feedback loop.
https://www.ti.com/lit/an/slyt612/slyt612.pdf
..although it is hard for me to tell with the crude IRIS schematic I have if they are including the CMOS switch in the amp feedback loop.
Last edited:
Maxim (now part of Analog Devices) has some good papers on this also.
Regarding your question, please see attached image captured from the Hafler 915 Preamp manual.
The distortion in the data sheets is worst case specified under certain conditions; so for less strenuous test conditions, the distortion will be less.
Seems that at least one way to minimize analog switch distortion is to reduce the voltage swing across the switch. Two ways to do that:
1. Reduce the current drawn through the switch Ron (for example, use a J-FET opamp/buffer after it)
2. Use the switch in the virtual ground input of an inverting op amp gain stage with the switch connected to the inverting input. Of course, you have to watch out for parasitic capacitance, etc. when you do this, but as you mentioned, you can also throw a switch in the feedback loop and play with the gain setting resistors to (mostly) cancel the switch's resistance.
mlloyd1
Regarding your question, please see attached image captured from the Hafler 915 Preamp manual.
The distortion in the data sheets is worst case specified under certain conditions; so for less strenuous test conditions, the distortion will be less.
Seems that at least one way to minimize analog switch distortion is to reduce the voltage swing across the switch. Two ways to do that:
1. Reduce the current drawn through the switch Ron (for example, use a J-FET opamp/buffer after it)
2. Use the switch in the virtual ground input of an inverting op amp gain stage with the switch connected to the inverting input. Of course, you have to watch out for parasitic capacitance, etc. when you do this, but as you mentioned, you can also throw a switch in the feedback loop and play with the gain setting resistors to (mostly) cancel the switch's resistance.
mlloyd1
