I now remember looking at a fr graph of the 105s many years ago. It rolled off fairly quickly at both ends of the spectrum at 20hz and 20khz. The comment was something like, " I would like to have seen response extended a bit more, especially at the lower end".
I now remember looking at a fr graph of the 105s many years ago. It rolled off fairly quickly at both ends of the spectrum at 20hz and 20khz. The comment was something like, " I would like to have seen response extended a bit more, especially at the lower end".
When the input is DC coupled there is no passive high pass filter on the input.
If there is some LF roll-off, then that is coming from some internal frequency shaping.
I am not an amplifier designer, but I believe it is bad practice to roll-off the frequency response inside the amplifier. It leaves the feedback affecting the performance. This is maybe what you are hearing.
Keep the amplifier as is, or sell it. Don't try to modify from B.Stuart's design philosophy.
If you have reason to want to agree with my non-designer intuition then select an amplifier with passive input filtering that has a passband that is wider than the input filtering and has adequate current delivery without the output voltage collapsing when speakers demand high currents. 6800uF in my experience is far too low for an extended "apparent" bass response.
I reported my Crimson experiments some years ago.
I bought the ce1100 kit and assembled it. I then spoke to the Designer and we agreed that doubling the smoothing capacitance would improve performance. I did that. I liked what I heard.
I then bought the parts to make my own version and this time I quadrupled the smoothing capacitance to +-6800uFx3 for each channel. Ths was the start of realising that skimping on the PSU affects performance, particularly bass performance. That ce1100 kit had just +-4700uF for each amplifier feeding 8ohms. I use >=+-20mF for all my 8ohms rated power amplifiers as a matter of course.
Last edited:
I agree this is lower supply capacitance than usual, 10,000uF-22,000uF per polarity would be more typical.
The nfb capacitor to ground is a little undersized as well, 220uF would set a more typical 2Hz LF rolloff.
As per post 19, the last thing I want to accomplish is to undermine or compromise what they do well right now. If that is the outcome, then I will not proceed. They are perfect except for bottom end weight which can be added otherwise. I have not yet heard an amp so incredibly fast/transparent.
You could add connections to the amps for the capacitors in the extra supplies.
Mount those supplies in an enclosure (no need to disconnect the rest of the
circuit in them). Then plug them in and have a listen to double the capacitance.
Not having heard these amps, part of the sound quality may be that they are inverting,
unlike most other power amplifiers. The input and feedback signals are directly combined
in the nfb network, instead of going to opposite sides of a diff amp input circuit.
Last edited:
Sure, but I'm not reading any responses with hardy approval. Mind you, I don't see any difficulty in connecting the additional power supplies, either fully or just additional capacitance. There's no harm in trying it out which I think I will do. I ran it by you all for your educated opinions and to make sure I don't blow anything up. I love overkill
They are complete in their own enclosures as they come in modular form, very elegant imo, patch chords from the amp enclosure to the psu enclosure. They are bound side by side with connecting bars underneath that serve also as feet/standoffs for heat dissipation.
Sure, but I'm not reading any responses with hardy approval. Mind you, I don't see any difficulty
in connecting the additional power supplies, either fully or just additional capacitance. There's no harm
in trying it out which I think I will do. I ran it by you all for your educated opinions and to make sure
I don't blow anything up. I love overkill
Use connectors (unlike banana plugs) that can't easily disconnect,
and check the polarity of the connections very carefully.
Make the output connections like those of the internal supplies.
Last edited:
I'd do some research on paralleling transformers if I were you, but I think you'll find that adequate increase in available current will be achieved by increasing the capacitance which also seems to be the consensus here. It will not affect the attributes of the amplifier that you like although it may seem to subjectively if it changes the tonal balance, ie. more bass/slam, whatever
I would not attempt paralleling the entire two supplies or the transformers, there's too much
uncertainty about sharing. Just parallel the capacitors.
The existing transformer should be good with twice the capacitance, so just jumper the extra
capacitors over to the amp. No other changes to the supplies are needed, since the rectifier
diodes will block DC from the unused transformer. Triple check all polarities.
If I were rebuilding these amps, each of the four supply modules would be configured so
each transformer parallels its dual secondaries for one output voltage, and twice the
capacitance per polarity, per amplifier. (If there are in fact dual secondaries on each transformer.)
Anyone have a power supply schematic?
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.