Well with the clean electricity one gets late at night along with good FM reception I've been listening to the TGM5 employing my 'tuned' up harmonic profile. It's the closest I've yet heard from a SS amp to what I heard from the SET mono block.
Well I'll stop there, builders pride always emphasizes the good in the sound of a new amp but I'm impressed with what I'm hearing. The sound is alive, but clean from bottom through mids to highs.
There is absolutely no turn on noise, nothing, nada, it's really quiet. No turn off noises either.
Nice to hear about your's experiences. Now you know what I have spoken about so many times in the SSA thread.
There are many different possibilites for one to adopt the SSA sound signature according to personal taste, also in a way as Esperado explained.
Regards, Andrej
Member
Joined 2009
Paid Member
too early to say, I have only been using the centre channel. I will look at hooking up for stereo this weekend.
I will use my Bryston as a reference for some further listening tests this weekend - it has a symmetric design based on complimentary LTP input stage. It's a supremely clean sounding amp to compare with.
Member
Joined 2009
Paid Member
Well, I finished wiring all the channels up - it's all finished except for the speaker protection. I decided to use some big electrolytics as temporary dc protection - only they didn't sound too good, in fact I felt like poking my ears with a sharp stick so I decided to take them back out and risk it for awhile.
I hooked up my YBA CD player as a source and connected up my floor standing PMCs.
The sound was not as good as yesterday. The 'tube' sound that was so enthralling with the small full-range speaker did not translate to the big floorstanders. I replaced the feedback resistors and returned it to a fully symmetrical design. This cleaned it back up. Now, on my PMCs it sounds really darn good. I'm not sure bass is as strong as the Bryston, but it's very close and overall I find it more musical. Where it seems to shine in particular (at least right now) is on the lower treble - where most of the harmonics appear for strings and female voice. Here the sound is simply gorgeous.
I also switched over my my B&W floorstanders. Sound is good, but after listening to the PMCs the B&W's just don't really cut it.
I'm still surprised how turn-on and turn-off is a non-event. You can cycle the power, turning it back on before the power supply caps are drained - not a sound, not a tick, you'd think it was simply on mute. I've never experienced such quiet with a DIY amp before.
The dc-offset appears to be rock steady too, which I'm still scratching my head over.
Well that's as much exposure as I want to give my PMCs without dc protection but I've heard enough to know this is a keeper. It's almost a waste to use it for my home theatre, but the enormous weight of the EI power transformers suggests that it stays where it is (and those darn things do have a noticeable hum to them).
TGM5 - so far I have to say I highly recommend it
yup.
I hooked up my YBA CD player as a source and connected up my floor standing PMCs.
The sound was not as good as yesterday. The 'tube' sound that was so enthralling with the small full-range speaker did not translate to the big floorstanders. I replaced the feedback resistors and returned it to a fully symmetrical design. This cleaned it back up. Now, on my PMCs it sounds really darn good. I'm not sure bass is as strong as the Bryston, but it's very close and overall I find it more musical. Where it seems to shine in particular (at least right now) is on the lower treble - where most of the harmonics appear for strings and female voice. Here the sound is simply gorgeous.
I also switched over my my B&W floorstanders. Sound is good, but after listening to the PMCs the B&W's just don't really cut it.
I'm still surprised how turn-on and turn-off is a non-event. You can cycle the power, turning it back on before the power supply caps are drained - not a sound, not a tick, you'd think it was simply on mute. I've never experienced such quiet with a DIY amp before.
The dc-offset appears to be rock steady too, which I'm still scratching my head over.
Well that's as much exposure as I want to give my PMCs without dc protection but I've heard enough to know this is a keeper. It's almost a waste to use it for my home theatre, but the enormous weight of the EI power transformers suggests that it stays where it is (and those darn things do have a noticeable hum to them).
TGM5 - so far I have to say I highly recommend it
yup.
Last edited:
My early impression of the less components 50W Latfet one I discussed in the main SSA thread as put together by a friend too, is that is a homogenous big sound, extra quick and smooth. Treble is really never solid state harsh, and overall detail level is very high but naturally spread. Bass full, elastic, pacey. I compared to my KT88 trioded 20W push-pull and although its power drive was naturally tamer for the doubled up 6x9 oval bass drivers of my 3 way than the SSA, it reminded me that this is an early amp, still need to see about things like more tonal differentiation on instruments, a bit more depth perspective, removal of some second harmonic or counteract what strikes me as some Mosfet mist-lack of tonal brilliance. Have to see about best bias for most suitable harmonics profile, squarewave best shape regarding subjective impact on overall tone. Then if not content, I will have to see about Dale resistors, other than Toshiba input stage BJTs, maybe a BJT output stage. I compared in my own system with vinyl as source. To put it in a nutshell my impression is this is maybe the best solid state diy platform to tamper with. Shows big potential, but its not a finished project yet. Smashing success on main lines, needs to build some finer musical character yet. Being class AB, it stands high between class A bias special amps on first strike.
Member
Joined 2009
Paid Member
I've not had much exposure to push-pull that isn't BJT output. I once had a MOSFET output home theatre amplifier but it sounded thin and lifeless (and had to be returned to the store). Plus there's something elegant in my mind about the CFP driver stage, I saw it first on Roenders amplifier and I remember people who built it liked it very much; I used it on TGM3 and the sound was very good.
I like your description of the bass as 'full and elastic' as I couldn't find the words to capture my feelings about it. But it is the treble that I like the most, not the high treble (which I still find is best done by tubes) but the lower treble where there's a bit more information anyhow - it seems to me to be really great - it's lively, detailed, and textured.
I hope you have some fun with your investigations !
now, I need to sort out the wire dressing, still have a bit of hum…
I like your description of the bass as 'full and elastic' as I couldn't find the words to capture my feelings about it. But it is the treble that I like the most, not the high treble (which I still find is best done by tubes) but the lower treble where there's a bit more information anyhow - it seems to me to be really great - it's lively, detailed, and textured.
I hope you have some fun with your investigations !
now, I need to sort out the wire dressing, still have a bit of hum…
Attachments
Noticed the smooth lower treble info galore too, its a Sziklai combo output, 1.2MHz -3dB. Sounds super full and big. I just think there is more to tune so to hit the tonality nail on its head too. Its a good amp, its a crime not to investigate fine tuning. Nice config your chassis pic. Congrats.
Did you try a 6Db/oct passive low pass filter at around 100KHz right before the input ?
Did you try symmetrical cables to feed your amp ? (shield and neutral together in the output, shield not connected input side).
Did-you try the good side's combination of all power plugs in your system to minimise power currents leakage between your different equiments ?
Did you provide proper HF filtering of the power cable, too ?
All that can change treble sonic quality.
Member
Joined 2009
Paid Member
It's not that there was anything 'wrong' with the high treble, in fact for the most part the treble is some of the best I've heard, especially the lower treble - just amazing. On the other hand, I have noticed that in the past I have preferred the sound of high treble from the few tube amps I've listened to when compared with SS amps, perhaps because in those examples the tube amps were a little more rolled-off. One tube amp however was not rolled-off and the output transformer not the best one could buy - the treble in that case was not good. I'm particularly sensitive to fatigue from treble, once I even experienced 'treble overload' at a live symphony (real instruments are darn loud!).
I have noticed that listening in the late evening the quality of the treble in all of my amps is better and I attribute this to the quality of the power. I haven't got any hf filtering in the power cable and this would be interesting to try when I have time - I am thinking of a simple common-mode choke. But for TGM5 I have used EI transformers which I understand are better at limiting hf ingress from poor power, and I have a CLC filter.
The r.f. signal filtering at the input to TGM5 is a 1k series with 1nF in parallel so it doesn't cut off that quickly.
Still, I think initial listening tests remain very positive and I will do more listening. I have to resolve some residual hum first and I want to get some dc speaker protection in place.
This evening the amp had it's first use in my HT system - the best my HT system has sounded since I build it. Strangely though, the main on-off switch has failed, it's stuck 'on'. I will have to replace it - the switch was bought new 3 yrs ago.
I have noticed that listening in the late evening the quality of the treble in all of my amps is better and I attribute this to the quality of the power. I haven't got any hf filtering in the power cable and this would be interesting to try when I have time - I am thinking of a simple common-mode choke. But for TGM5 I have used EI transformers which I understand are better at limiting hf ingress from poor power, and I have a CLC filter.
The r.f. signal filtering at the input to TGM5 is a 1k series with 1nF in parallel so it doesn't cut off that quickly.
Still, I think initial listening tests remain very positive and I will do more listening. I have to resolve some residual hum first and I want to get some dc speaker protection in place.
This evening the amp had it's first use in my HT system - the best my HT system has sounded since I build it. Strangely though, the main on-off switch has failed, it's stuck 'on'. I will have to replace it - the switch was bought new 3 yrs ago.
Last edited:
The problem is often on the recording side. Mikes with peaks in treble (Shoeps or Newman), corrections added by the sound engineer (treble are hifi) etc...
Most of the recordings have too much treble. On the other side, many enclosures, and for the same reasons, emphasis the treble too.
Last, there is not so important informations after 16Kz, if you get dynamic enclosures (with real energy in treble, and no sound compression, means not those "pssss pssss" of dome tweeters).
On my side, i use a horn for mids and treble p, my two wzays system, (1" JBL) It do not goes very hight in frequency (14/16Khz) but with a real and strong energy and that the only way i have found to get musical instruments natural to my ears. Adding a tweeter can be impressive, but awfully far from real. And i get tired very fast.
I"m a little like you, it seems.
But try my cable schematic for signal, with a good professional symetric microphone cable, you will be impressed by the add of transparency and *natural* details, and less aggressive high trebles and never, never use special expensive high end magic loudspeakers cables ;-)
My two cents.
Hi guys 
I am glad that TGM5 will be a good HT amplifier as it was intended to be. It is as you've said even overkill for that purpose, I agree. It was designed for that purpose so I didn't want to interfere too much.
What I would do in a amp for a stereo system would be VAS cascodes, T5&T6 super beta BJT's, 100ohm/100pF input low pass filter and larger uF filtering caps from Mundorf or Panasonic.
My
I am glad that TGM5 will be a good HT amplifier as it was intended to be. It is as you've said even overkill for that purpose, I agree. It was designed for that purpose so I didn't want to interfere too much.
What I would do in a amp for a stereo system would be VAS cascodes, T5&T6 super beta BJT's, 100ohm/100pF input low pass filter and larger uF filtering caps from Mundorf or Panasonic.
My
Hi Esperado
Your presence is golden since you're the man with endless experiences. I am so glad I woke you up from years of DIY absence.
Best regards, Andrej
You are TOO nice.
I can dream to have both a shorter experience (Who is this old man in the mirror ?)... and your talent.
Regarding your 100ohm/100pF low pass filter proposal, i would agree if the preamp is able to deliver full power under 100ohm at 3Mhz...
The low pass filter is not so much against Hf rejection, but to reduce the slew rate in a matter that any signal applied to the amp is *a lot* slower than the amp's speed, OMOH.
It is amazing how important too is the power ability of a good preamp. I made mine able to drive a 10ohms load at nominal level and charged at 1K, half inside, half in the amp side.
I use a video current feedback IC with a discreet fet buffer in the CR loop. 100ohms in serial, 2.2K in // ( internal charge ).
On the amp side, 1K in serial, and 10k to the ground for the 0V reference, with as a low pass cap in //, 22pf for benchs,.
For normal use 10k replaced by 1K, and 470pf (tuned with my ears ;-). Not standard at all.
Hi Esperado
First thanks for the compliment and please don't be too modest, the way you upgraded legendary Elector's Crescendo is for the history records.
Just a little comment regarding 100ohm/100pF input low pas filter I proposed.
100 pF capacitor has 530,52 ohm impedance at 3 MHz plus serial 100 ohm resistor, resulting in a 630,52 ohm of an amplifier input impedance. This filter was chosen because of two reasons, first, it is matched with amplifier power bandwidth, exactly following the gain slope, meaning we don't loose any of the amp's speed performance, second, it successfully reduces unwanted HF signals (exceeding amps frequency response) which would tend to stimulate an amplifier to oscillate.
We also have to consider preamp's output impedance plus potential serial resistor at pream's output (usually used) so to my opinion 1 k/1 nF is simply too high value for an input low pass filter - capacitor.
Member
Joined 2009
Paid Member
I'd better understand properly what you are suggesting. For my input signal cables I use single core coax - I got this coax from some recycled equipment and it's designed for high frequencies (MHz), I've not tried to match or terminate the characteristic impedance. The wiring of the 3 channels was tricky, poin-to-point on the Alps potentiometer was a trying effort to get all the cables soldered up properly - I am very reluctant to make any changes and may want to try your trick on another project first. Anyhow, can you make a simple drawing to teach me what you mean ?
I'm very interested to hear you mention horns. I know it's off-topic but since this my thread I can take liberties
My TGM5 was designed for big 3-way floorstanders and hence I wanted a negative feedback amp based on sampling the voltage at the output - because this is what the designer of the speakers had in mind. But for my 'main listening pleasure' upstairs (not for my home theatre) I am planning something different - and the front runner is a SS amp driving below 100Hz, a PP tube amp driving low-mid horn up to 400Hz and a SE tube amp driving a horn above 400Hz. What is your thoughts ?Hi guys
I am glad that TGM5 will be a good HT amplifier as it was intended to be. It is as you've said even overkill for that purpose, I agree. It was designed for that purpose so I didn't want to interfere too much.
What I would do in a amp for a stereo system would be VAS cascodes, T5&T6 super beta BJT's, 100ohm/100pF input low pass filter and larger uF filtering caps from Mundorf or Panasonic.
My
Yes, building TGM5 was really an interesting project, the HT was really just an excuse to do it
On the other hand, I think people underestimate the benefit of hi-quality sound for HT as there are not only some movies that are really made for good sound (e.g. Master and Commander) but also some blu-ray mutli-channel music discs. When you have good sound for music, it's intolerable to have bad sound for movies.I agree, the next step is to cascode the VAS, taking the heat of the VAS allows higher performance parts to be selected, more linearity to be achieved. Not sure about super-beta devices. Maybe try FETs as Hugh and others like PMA have suggested.
With gobs of nfb the main rail caps are less concerning to me than the filters on the front end of the amp - and the royal way might be to use a shunt regulator there ?
Interesting comment - I have never designed a pre-amp for my amplifiers. I have additional secondary windings on the power trafo's but the chasis is getting a little crowded. I'm not sure if this is worth the effort.
Unless i did miss something , the mods you did to the crescendo are not
really reproducible , as the standing currents seems either badly defined,
and as such are active components dependants, or in some case
can be simply absent.
Please, forgive-me, but i think you miss something, or you are living on a planet where Ohm law has been writen by a federal court?
I do not understand one word of what you are talking about. Did-you know why, in low signal Hf amps, so high currents and low impedances are used ? Please, use the sim and look to the current of each stage. or better, build-it in the real world ;-)
@lazycat: As i said, i tuned-it by ears, and for my system. If i take care, since decades, to build very fast amps and preamp (slew rates), i do not worry too much about "paper" performances concerning the bandwidth. It would be long to explain. Coherency between slewrates and bandwitch with digital sources, power handling capabilities at high frequencies, dynamic IM...
I remember, when i was managing a big concerts PA system company, how surprising was the lack of treble, when we used some Nakamichi K7 tapes, you know, at low level, before and after the show: no real treble, dead sound, they where given to be flat up to 22kz. But at -10db ;-).
Depending of your system, and after you had made sure that the previous stage will be able to easy drive the charge, just try to play with the filter to reduce the high bandwitch, blind, and find the values witch give the most realistic sound on a neutral very dynamic recording. I bet you will be surprised by what you will prefer and measure ? And it can be very different from a loudspeaker to an other.
One thing is sure, you can feel differences between a 200V/µs amp and a 1000V/µs one, even when you filter the incoming bandwidth at 100Kz -3db. Specially on bass drums kits piano and body of crash cymbals. Filtering the bandwitch with a passive network do not kill the quality offered by a fast amp.
@Bigun: I cannot talk about your choices, listening pleasure VS analytic reproduction, tastes, habits, culture can be so different between people. We are all looking for something different. Because i was a sound engineer, i am looking for transparent systems able to reveal all mistakes in recordings and mixes. Not always the most agreeable listening experience. Reason why i avoid tube amps, on my side. Now, i'm out of the business, i should change my point of view, I suppose ?
About cabling, very simple. Use a symetrical shielded cable. red wire on the +signal out. black and shield connected together to the ground of your RCA out plug. Right ? On the other side the (RCA witch is connected to the amp), isolate the shied for no current flows in it, and just use the red for + and black for RCA ground. If you can add a ferrite near the output side, it will be better. I would like to add that parasitic HF has worse effects in the CR of any output stage than in the input of the following one.
Just my points of view, but based on thousand experiences on many various systems and locations, in studios or big PA systems.
Member
Joined 2009
Paid Member
I understand, you are talking about the interconnects and what you propose also cuts ground loops between interconnected equipment. I don't have cable suitable for this experiment but I've added this to my 'mental list' of things to try. I am thinking that some similar benefits may apply to the use of an input transformer at the amp side - does this make sense ?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.