UK Vintage Radio Repair and Restoration Discussion Forum

UK Vintage Radio Repair and Restoration Discussion Forum (https://www.vintage-radio.net/forum/index.php)
-   Vintage Audio (record players, hi-fi etc) (https://www.vintage-radio.net/forum/forumdisplay.php?f=14)
-   -   Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement. (https://www.vintage-radio.net/forum/showthread.php?t=168596)

bikerhifinut 2nd Jul 2020 2:28 pm

Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
I am about ready to start doing a cap replacement on a Rogers ravensbrook Series2 amplifier.

I carefully powered it up and did a quick set of voltage measurements around the PSU side of things. I noted a small amount of 100Hz hum but as its basically down to a bare PCB just now and the mains transformer etc doesn't have a mains earthed steel case around it I think some hum could be to insufficient earthing. So careful when poking the meter probe in. And of course the capacitors are pretty old, most having been there since 1971.

So I did a quick voltage check, First with the mains voltage setting at 240V (which would be its usual setting) and again quickly at 220V to check how the regulation was doing.
The regulator is a simple one, consisting of a 43V 1W zener diode, bypassed by a 120uF capacitor to the base of a PNP Germanium power transistor which has its collector on the negative side of thee rectifier and emitter outputting to the signal earth power rail. Theres a 390 ohm resistor between base and collector. The main reservoir capacitor is a weedy by modern standards 1000 uF but memory serves me that increasing this value causes fairly serious issues and I feel that the regulation is the main hum "filter".
The meter shows that the regulation seems fine as it stays stable at just less than 43V on the 240V tap and a shade above at 44.1V on the 220V tap. No signal and a pair of 8 0hm speakers connected. This puzzles me a bit as I thought the zener and pass transistor would hold the output voltage steady at whatever input voltage. Insight here would be appreciated, keep it simple please as I can be hard of understanding.

The pass transistor should be a 2N3611, this blew many years ago and the best germanium power transistor I could lay hands on (or rather my dad who blagged it off an instrument tiffy at his work) was a proper OC28. The spec isnt quite as good as a 2N3611 in terms of power handling but its rated at 6 times the maximum current it's expected to pass and as far as I can see it just about copes with the voltages its exposed to. It hasnt blown yet but I am wondering if the low level background 100Hz could be down to it breaking down a bit? If so I wondered if any decent PNP TO3 transistor would be ok including silicon, I am assuming that good PNP silicon power transistors werent readily (cheaply) available when this amp was in production. I havent looked for a 2N3611 as yet.

So the plan is to firstly replace just the PSU caps and power supply decoupling caps and then give it a listen and maybe bung the scope on it as a check.

Would it be wise to replace the speaker coupling caps too at this point? I would imagine after 49 years they will be pretty dried up.

I'll replace the rest of the electrolytics in stages once the power supply is sorted out. I am very aware of the hazards of replacing all the caps at once and then facing the possiblity of tracing a fault not knowing where I boo booed!

On the face of things this isnt a major resuscitation from a non working state, luckily, but this particular amp has great sentimental value for myself and Father and that's why I am going very carefully.

I hope it's not cheeky or rude but can we keep the comments and replies focussed on my queries or relevant to each stage of the refurbishment please? I can also assure everyone that I have sourced a full complement of electrolytics at or as close as current preferred values allow.

Andy.

turretslug 2nd Jul 2020 3:40 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
Zener diodes aren't "perfect" devices, they do have some slope resistance, i.e. changing the current through them will result in a slight change in the voltage across them. As the transformer secondary voltage rises as a result of feeding it on the 220V primary as opposed to the 240V option, so the current through the Zener increases and the voltage across it rises slightly.

Radio Wrangler 2nd Jul 2020 4:43 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
There doesn't *have* to be a relationship between speaker coupling capacitor in general terms.

An amplifier can give an output to far below the 100Hz ripple frequency on a reservoir provided the amplifier is under feedback control. After all, DC power supplies and amplifiers do work. You can consider the output capacitor to be infinite!

It comes down to preventing problems during turn-on and turn-of processes. Catching diodes in the right places and careful choice of ratings will do the job.

Upping the value of a reservoir will increase the ripple current amplitude while reducing its on-time and will also act to magnify rectifier losses. Rectifier ratings need checking but otherwise you can get away with quite a large increase.

So in an amplifier being repaired, some increase is reasonable, beyond that you need to do some calculations and check ratings.

David

bikerhifinut 2nd Jul 2020 4:59 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
Thanks.
The point about the reservoir capacitor is that as a youth I tried a larger value when I fixed the original PSU fault when the pass transistor packed in under an accidental overload condition, The 1A protection fuse wasn't fast enough.
All I got was a boomy overblown bass and a tendency to blow the fuse when the wick was turned up not much above a moderate level on transients I guess due to larger reservoirs ability to deliver instantaneous current for a short period of time.
So I have stuck with 1000uF as per the original design. everything else so far has been as near as I can get to the originals in modern preferred values.
I only wondered about the speaker output coupling caps because I was a bit concerned but I imagine if they are dried up a bit, and I did discover a couple of bulging caps, then they'd go high on the ESR and likely not present as low a load. More a worry if they go S/C?
Andy.

bikerhifinut 2nd Jul 2020 5:00 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
:thumbsup::thumbsup:
Quote:

Originally Posted by turretslug (Post 1265855)
Zener diodes aren't "perfect" devices, they do have some slope resistance, i.e. changing the current through them will result in a slight change in the voltage across them. As the transformer secondary voltage rises as a result of feeding it on the 220V primary as opposed to the 240V option, so the current through the Zener increases and the voltage across it rises slightly.

:thumbsup:

bikerhifinut 2nd Jul 2020 6:11 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
So far so good............
All PSU and decoupling caps replaced and its working fine, also then replaced the output caps and then the coupling caps to the driver transformer primary again all checks out. Any hum I am putting down to a lack of earthing and screening as its out of the steel chassis still.
I'm starting on the coupling caps now as I measured a DC voltage of about 11V on the input capacitors on the disc stage (6uF@16V), I am using 6,8uF here.
Theres a few emitter bypass capacitors that are very low working Voltage, 6.4V and one at 3V according to the manual but the one thats fitted is a 6.4V. This is on the driver stage and I havent a 200uF or 220 at such a low voltage. I do have a number of 100 uF at 16V and wondered if a couple of those in parallel would be ok. Likewise the 47uF bypass caps I only have 63V in the spares bin but unlike plastic films will there be enough polarising voltage to keep them formed? Before I dive in.
I have a notion that old electrolytics may be a reason why a lot of older gear gets a reputation for having a less good sound quality, given everything else has stayed in spec.
I'm prepared to buy some low voltage capacitors here if necessary. But if the ones I have will do the job then thats good.

A.

thermionic 2nd Jul 2020 7:03 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
As others have said, slightly increasing the values of reservoir and output coupling capacitors will be beneficial to the amplifier.

Even in the early 1970’s, larger value electrolytic capacitors were bulky and expensive, so manufacturers were keen to keep the capacitances to the minimum they could get away with.

Today we are spoilt with the availability of high capacitance, compact components at reasonable prices. However, if you want to keep the amplifier sounding as it would have done originally, then sticking with, or very close to, the original values is the way to go!

Cheers.

SimonT.

bikerhifinut 2nd Jul 2020 10:24 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
Quote:

Originally Posted by thermionic (Post 1265958)
As others have said, slightly increasing the values of reservoir and output coupling capacitors will be beneficial to the amplifier.

Even in the early 1970’s, larger value electrolytic capacitors were bulky and expensive, so manufacturers were keen to keep the capacitances to the minimum they could get away with.

Today we are spoilt with the availability of high capacitance, compact components at reasonable prices. However, if you want to keep the amplifier sounding as it would have done originally, then sticking with, or very close to, the original values is the way to go!

Cheers.

SimonT.

Thanks Simon.
So can anyone tell me other than my previously considered reasons, why increasing reservoir capacitance causes problems in terms of excess bass and blowing the 1A supply protection fuse if the wicks turned up. Which is why I kept to 1000uF main reservoir?
I checked the Voltages. to the nearest Volt as it varies slightly as the mains drifts.
off rectifier across reservoir capacitor 47V
Across zener 43V
between base and collector of pass transistor 4V
between emitter and collector of pass transistor 4V
and from the positive to negative of main supply 43V
So I figure its working OK? Or have I missed a fundamental?
if my maths is right then the 4V drop across the 390 ohm resistor between Base and collector means theres about 10mA flowing through the zener and thus a dissipation of 41 mW which is well inside the diodes 1W rating and the resistors 1/2W.
I've so far also replaced the coupling caps on the disc stage and output driver, and speaker output caps. but nothing else yet.
I feel that its got to be worth working through them all though after 50 years, only the electrolytics though as all the rest are standard Mullard polyesters etc which should still be good.
A.

bikerhifinut 3rd Jul 2020 11:54 am

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
Well I have got it running nicely from a line level source into a pair of High quality speakers in the main room.
So far all I have replaced are all the electrolytics in the main PSU, and associated decoupling capacitors between stages.
also the output capacitors, now 2200uF instead of 1500 and the 640uF coupling capacitor to the output stage driver transformer (680uF used).
I renewed the input coupling caps on the disc stage too. The disc stage is still showing instability which I hear when the volume is turned up. (Oscillation and motorboating) although all decoupling capacitors have been replaced. This needs investigation, the only electrolytics left are a pair of 47uF emitter bypass electrolytics. I think it warrants it's own thread.
I havent touched any others like the emitter bypass caps and the many small interstage coupling capacitors, there's no DC coupling at all in this old lady.
I'm wondering if its worth replacing them all on a reliability basis, and would it improve the sound any? It's sounding pretty good I reckon but I'd describe the overall sound as a bit "muddy" in the mid band compared with the excellent marantz PM45 of my sisters that's still being stored here due to a thwarted house move. So I am thinking this is the difference between the early 70s/late 60s and the mid 80s when silicon amplifier technology was probably at its apogee. I dont think there were any seismic advances after that, more refinements than anything else.
Whats the experienced restorers opinion? I have the capacitors in stock. They wont go wrong for future jobs, as all are popular values in the 10 uf and 47uF values.
Whilst the amps out of its casework there's about a couple of hours work to swap out the remaining electrolytics.
Thanks in advance for your thoughts and patience with me Ladies and gents.

A.

bikerhifinut 3rd Jul 2020 2:49 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
Phew!
Think I may have got to the instability on disc.
The only electrolytics I hadnt replaced were the emitter bypass on the second stage of the disc preamp. All other capacitors were sound looking ceramics or polyester "tropical fish" or the tubular 5% jobs in the RIAA EQ.
So I swapped the electrolytics out, they didn't look to be leaky or even swollen but as they have a very low voltage on them I doubt I would anyway.
A quick hook up to the turntable and speakers and all seems good. A trifle noisy on the 50Hz hum but its still a bare PCB with no path to mains earth until the steel cabinet is back in place. Its also slightly noisy in terms of hiss etc but probably as good as it got in 1970 (No metal film resistors). Anyway I cranked the volume up and the motorboating and hf "Birdies" Had gone. I never realised that a bypass capacitor would have so much effect. Expert explanations will be very gratefully read.
So its concentrated my mind now to replace the lot as if ones gone then its a fair bet there are others about to or already failing.
I'm actually chuffed to bits that the old amp is doing such a good job in a modern audio set up. Father and I always liked the way the Rogers made music, no dramas, no flashing lights, but always pleasant. And a nice reminder at times that New or "Modern" isnt always that much of an advance in terms of grin factor!
These amps are a lot nicer to work on than their Armstrong and Leak contemporaries in my inexpert view, that nice Glassfibre board with well spaced and neat component layout makes it easy to follow and work on. Just don't use too hot an iron desoldering as the tracks will lift if not careful!

Back to the bench.

Andy.

thermionic 3rd Jul 2020 6:57 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
Sounds like you’ve nailed it, Andy!

What I’ve found with these old AC coupled amps, is that whilst the interstage capacitors might check ok for ESR and capacity, they can be overly DC leaky. This can upset the biasing slightly in the following stage, which can result in greater distortion and other odd effects.

Cheers.

SimonT.

bikerhifinut 3rd Jul 2020 11:02 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
Thanks Simon.
I swapped out the entire electrolytic contingent, only mildly fiddly.
There were some very low working voltages but I made do with as near as I could get. There were a couple of 220uF at 3V specced! there was a 200 at 6V on the board and I only had 220uF at 63V so I found some 100 x 16V lurking in the stock, basically paralleled them up to make a pair of 200uF. Not ideal but it works. I figure the others will form ok and last me out.
But the end result was, I think, remarkable. It could be psychological of course, and I wouldn't deny anyone suggesting it, but I am sure the sound has opened out and sounds quite "modern" whereas it was a bit "fluffy" and rolled off previously.
All things considered it's a very pleasant thing and if 10W is enough for you (it is for me in a smallish room) it works very well with more up to date turntable and cartridges and didn't shame itself fed CD from a very good player. The more recent disc stages I own will better it, but that's not to denigrate what to me is a creditable performance.
I hope my old Dad can appreciate it when he gets it back next week on his 90th birthday. I shall slot it in between the PL12D and Wharfedale Dentons for a nostalgic experience.

Andy.

Radio Wrangler 3rd Jul 2020 11:56 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
Anything you've fixed yourself sounds 76.8215% better than anything bought in any shop.

Anything you've built yourself sounds 135.718% better than anything bought in any shop.

Anything you've designed yourself sounds 229.8333..% better than anything commercially available.

These are constants of our universe.

David

fetteler 4th Jul 2020 12:15 am

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
Spot on David :-D

Steve.

bikerhifinut 4th Jul 2020 11:06 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
I spoke too soon.
The amp seems to be working just fine and dandy and sounding like it should.
The issue is that the "slope" filter now seems inoperative.
Its described as an adjustable low pass filter working above 6kHz, variable from zero to 20dB per octave above this frequency. it looks like a complicated treble cut network inserted between the first 2 gain stages. What I can't understand is if I had made a mistake, it would be unlikely to affect both channels equally and as it's in the signal path with 2 capacitors that I replaced then errors there would result in no sound.
I shall have to look at it but without any actual service data all I can do is check voltages around the active devices.

Herald1360 7th Jul 2020 9:22 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
How's your hearing? If you can't hear anything above 6kHz anyway, how steep the slope is on the filter is academic!

Have you got an audio frequency sig gen? Oscilloscope?

bikerhifinut 8th Jul 2020 3:27 pm

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
That is a very apposite question!
Actually my hearing is pretty good and I can easily hear well above 6kHz still.
I had the benefit of an employer who gave us all hearing tests annually, I worked in noisy environments for substantial proportions of the working day so hearing protection has always been high on my priorities, something that some younger folks should be aware of. Anyway at my last test before I retired the Nurse insisted on a rerun as she thought I was cheating! Nuff said.
But you are correct on the subtlety of the slope filter.
It is in fact working today I can hear the effect as I turn the pot and its more marked as I move back to the listening position rather than between the speakers, I suspect directionality of HF sound from the tweeters is playing a part.
It is extremely subtle but surprisingly effective on some older records with a bit of surface hash etc so its doing its job.
I think when I did a complete replacement of the electrolytics, Imust have restored the filters to their as new performance. Theres a couple of electrolytic coupling caps and also one in the local feedback network around the slope circuit so my suspicion is as one or more of these drifted off spec it interacted with the fiter network.
I don't subscribe to the concept of "burn in" of new components that's best left to another well known thread on here, but I did wonder if the new electrolytics needed a bit of time to "reform" as some will not have had very much of a polarising voltage applied to them.
I'm very relieved that I haven't had to dismantle the case to perform extra surgery!
Thanks for the input Chris, it does highlight just how subtle the Rogers Slope filter is and actually how unintrusive it is to the main musical enjoyment of the amp. I'm finding I rather like it very much notwithstanding its "old fashioned" design using a driver transformer on the output pair.

Andy.

Synchrodyne 9th Jul 2020 6:04 am

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
1 Attachment(s)
As far as I know, the transformer drive was a deliberate choice by Rogers, and not simply a carryover of older practice. It was first used on the Ravensbourne amplifier of 1967 and then scaled down for the Ravensbrook of 1969 (?). The standard Lin quasi-complementary output circuit had known shortcomings that exacerbated crossover distortion, and in 1967, the fully complementary alternative was hampered by the limited availability – and probably higher cost – of suitably large PNP output transistors, and those available tended to be rather fragile. The “fixes” for quasi-complementary circuit were still in the future, the very near future for the Baxandall/Quad output triples, and 1969 for the Shaw and Baxandall diodes. Thus transformer drive, with careful design and manufacture of the transformer, was in 1967 a reasonable option if the objective were to reduce crossover distortion at its source.

It is interesting to note that Bailey chose transformer drive for his 1966 amplifier design (Wireless World (WW) 1966 November p.542ff). His reasons for so doing were more fully articulated in the article on his 1968 fully-complementary amplifier (WW 1968 May p.94ff.) I have attached the first page of the article, and here is a salient quote therefrom:

“Compared with the quasi-complementary amplifier, the transformer-driven amplifier has the great advantage that the input impedances to the two sides of the output circuit are identical. This means that if a suitable quiescent current is used in the output transistors, cross-over distortion will be almost completely absent.”

I suspect that if you did a plot of distortion vs, power output, the Rogers transformer drive circuit would show decreasing distortion with decreasing power right down to the lowest levels, whereas amplifiers using the standard quasi-complementary output (without any of the “fixes”), and so relying on heavy negative feedback to achieve the full output distortion specification numbers, would more likely show increasing distortion at very low power levels.

Notwithstanding Bailey’s assertion that the fully complementary output was “the answer”, there were some dissenters, notably Vereker of Naim, who argued that there was in fact departure from symmetry at higher frequencies. As far as I know Naim used the quasi-complementary/Baxandall diode output in its early power amplifiers.


Cheers,

Radio Wrangler 9th Jul 2020 6:21 am

Re: Rogers Ravensbrook series2 amp. Before attempting a capacitor replacement.
 
I was surprised that if an interstage transformer was used in a transistor amplifier, the opportunity wasn't taken to use a significant step-down ratio (voltage) in the transformer and to drive the output transistors in common-base mode. There's an advantage in linearity and bandwidth.

Stage gain would have been less, so I suppose the cost of more stages killed it.

David


All times are GMT +1. The time now is 7:53 am.

Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.