That should be great. Keep the heatsink temp at about 50deg. C
I would use a bolts rather than a clip to secure the devices.
Jam
I would use a bolts rather than a clip to secure the devices.
Jam
Could the outputs be bolted to a reasonable size aluminum chassis?
Is it about 2.4W each for the four output devices?
Is it about 2.4W each for the four output devices?
Hi Kozard,
That would be fine as long as the devices are properly insulated from the case and you use heatsink compound and or a silicon pad.
With 24v rails I would bias the amp for about 100 to 125 mA.
Jam
That would be fine as long as the devices are properly insulated from the case and you use heatsink compound and or a silicon pad.
With 24v rails I would bias the amp for about 100 to 125 mA.
Jam
Last edited:
Heres my iteration of this wonderful circuit. Its for a dual mono modular preamp. I cascoded the mirrors with extra pair of bjts, all of em being bc337 327. The jfets are j113 175 and the mosfets 9610/610. Also i used the optcoupler bias circuit borrowed from the whammy headphone amp. I think for whatever extra distortion the optocoupler adds, the benefit of safely running higher bias is immense. Also added caps to drive transistors to slow them down a bit. Its probably not doing proper looking squarewaves anymore and instead rounding them off but the modules sound beautiful. Its a keeper for sure
Attachments
Donovas,
Nice work. I must have take some work to get matched jfets . I would suggest
you try the j74 and k170 and compare results. The idss for the chosen devices was about 8mils but if you have higher values you can always adjust the mirror to compensate.
I am working on a bipolar output stage for my next design.
Jam
Nice work. I must have take some work to get matched jfets . I would suggest
you try the j74 and k170 and compare results. The idss for the chosen devices was about 8mils but if you have higher values you can always adjust the mirror to compensate.
I am working on a bipolar output stage for my next design.
Jam
Jam,
Which headphones would you recommend pairing with the HPA1?
LCD-X is outside of my budget right now. I currently am using ATH-a900x (Audio-Technica) although I might prefer the sound of my older 900 or 700 to the 900x.
If something like the LCD-X is much more suitable then what about the Fostex T50RP MKIII Headphone? (Planar but more budget compatible.)
Which headphones would you recommend pairing with the HPA1?
LCD-X is outside of my budget right now. I currently am using ATH-a900x (Audio-Technica) although I might prefer the sound of my older 900 or 700 to the 900x.
If something like the LCD-X is much more suitable then what about the Fostex T50RP MKIII Headphone? (Planar but more budget compatible.)
Hi kozard,
The LCD-2 Classic (If you can find a deal) and Focal Listen Pro is pretty good at the price.
Hi-Fi Man have a few open back units that are fairly good I recommend you try the HE400SE . My favorite at the price point is the Sundara Headphones & portable audio - HIFIMAN.com
The Fostex is pretty good but not as refined.
I am partial to open back headphones and I would highly recommend the LCD-2 if you can swing it. The HP-1 was designbed using the LCD-X and LCD-3 and a Senheiser 650.
Regards
Jam
The LCD-2 Classic (If you can find a deal) and Focal Listen Pro is pretty good at the price.
Hi-Fi Man have a few open back units that are fairly good I recommend you try the HE400SE . My favorite at the price point is the Sundara Headphones & portable audio - HIFIMAN.com
The Fostex is pretty good but not as refined.
I am partial to open back headphones and I would highly recommend the LCD-2 if you can swing it. The HP-1 was designbed using the LCD-X and LCD-3 and a Senheiser 650.
Regards
Jam
Last edited:
The LCD-2 sound fantastic. But they're also bricks.
I found an Ether Flow 1.1 used for similar money. Do it sound as good? No. But it is close, and it's not a brick.
I use the LCD for shorter sessions and the Ether for longer ones.
I found an Ether Flow 1.1 used for similar money. Do it sound as good? No. But it is close, and it's not a brick.
I use the LCD for shorter sessions and the Ether for longer ones.
……on a lighter note.
Attachments
Last edited:
I have an LCD-3 with some comfort tweaks. I am mostly listening to my speakers these days, so if someone is interested in getting some LCDs, send me a PM.
The FQP3N30 are back ordered at this time (Mouser). I was thinking of trying IRF9610 and IRF710 temporarily until I could get the FQP3N30 and FQP3P20. (From Digikey searches I am a bit worried that many FQP devices are being obsoleted.)
However when the IRF9610 arrived their Rdson and capacitance indicate that they are a much larger device (IRF9640 size) labeled as IRF9610. The IRF710 measure normally. The same order had IRF9610 and IRF9640. The IRF9640 were for a different project and the IRF9610 and IRF9640 measure exactly the same. That seller was ok in the past for other parts but you need to always watch out.
I have IRF9640 & IRF640 in my parts bin but I assume that the device size, Idss and capacitance are far too high for even temporary usage in the HPA-1?
My PCB have shipped so I should have them in about 3 weeks.
However when the IRF9610 arrived their Rdson and capacitance indicate that they are a much larger device (IRF9640 size) labeled as IRF9610. The IRF710 measure normally. The same order had IRF9610 and IRF9640. The IRF9640 were for a different project and the IRF9610 and IRF9640 measure exactly the same. That seller was ok in the past for other parts but you need to always watch out.
I have IRF9640 & IRF640 in my parts bin but I assume that the device size, Idss and capacitance are far too high for even temporary usage in the HPA-1?
My PCB have shipped so I should have them in about 3 weeks.
kozard,
The IR p-channel devices exhibit a non-linearity (well documented), and
even though this effect is less apparent when used as followers I would use the Fairchild devices if possible.
Temporarily the IR devices should work fine, there is enough current (18 mils) to drive the IR's. You might loose some hf response and have slightly higher distortion but no problem. Increase gate resistor values if you notice any instability.
Jam
The IR p-channel devices exhibit a non-linearity (well documented), and
even though this effect is less apparent when used as followers I would use the Fairchild devices if possible.
Temporarily the IR devices should work fine, there is enough current (18 mils) to drive the IR's. You might loose some hf response and have slightly higher distortion but no problem. Increase gate resistor values if you notice any instability.
Jam
FQP3N30, 1336 piece in stock :
https://www.newark.com/on-semicondu...-ch-300v-3-2a-150deg-c/dp/84Y9967?ost=fqp3n30
200 pcs in tubes of 50:
N-Channel MOSFET, 3.2 A, 300 V, 3-Pin TO-220 onsemi FQP3N30 | RS Components
Patrick
https://www.newark.com/on-semicondu...-ch-300v-3-2a-150deg-c/dp/84Y9967?ost=fqp3n30
200 pcs in tubes of 50:
N-Channel MOSFET, 3.2 A, 300 V, 3-Pin TO-220 onsemi FQP3N30 | RS Components
Patrick
Ordering what was in stock I ended up ordering FQPF3N25 and FQP3P20 last night at Mouser. Not quite the same as FQP3N30 but hopefully ok? [I had to place a Mouser order for other components so I tried to order something close with the same FedEx charge.]
If that doesn't work I will have to try Newark. Digikey does not have them.
Also FYI the HIFIMAN HE400se is on sale (sale started yesterday) at $129 at "hifimanofficial" on eBay. If I like them I can later save up to get the HIFIMAN SUNDARA in the future.
If that doesn't work I will have to try Newark. Digikey does not have them.
Also FYI the HIFIMAN HE400se is on sale (sale started yesterday) at $129 at "hifimanofficial" on eBay. If I like them I can later save up to get the HIFIMAN SUNDARA in the future.
...I am working on a bipolar output stage for my next design.
So I've heard 😀
kozard,
The IR p-channel devices exhibit a non-linearity (well documented), and
even though this effect is less apparent when used as followers I would use the Fairchild devices if possible.
Temporarily the IR devices should work fine, there is enough current (18 mils) to drive the IR's. You might loose some hf response and have slightly higher distortion but no problem. Increase gate resistor values if you notice any instability.
Jam
Eventually I was able to order the right n & p Q FETs from Arrow. (I also discovered they have free shipping for $50+ orders.)
I have the KSC1845/KSA992 from Mouser but I was wondering if you could comment on the suitability of other types. Specifically I have lots of used genuine 2SC2240/2SA970. Are the KSC1845/KSA992 better? This is really just a theoretical question for my own understanding as I do have the KSC1845/KSA992 already. So far I am led to believe that the KSC1845/KSA992 are smaller/lower current/lower capacitance but otherwise similar to the 2SC2240/2SA970. I am also curious as I have a large quantity of generic 2SC2240/2SA970 and a lower priority curve tracer project. Right now my curve tracer only looks at the low voltage and low current region but at least in that region the clones look very similar. Later I plan to add amplifiers/buffers to extend the voltage and current range of the curve tracer.
Finally, why use a high voltage transistor like the KSC1845/KSA992 as opposed to something like the BC550C/BC560C or BC546B/BC556B? Again I will use the genuine KSC1845/KSA992 I just want to understand more about choosing the best type for this sort of design.
- Home
- Amplifiers
- Pass Labs
- Pass HPA-1, what do we know?