|
Vintage Audio (record players, hi-fi etc) Amplifiers, speakers, gramophones and other audio equipment. |
|
Thread Tools |
28th Jun 2017, 7:14 pm | #1 |
Hexode
Join Date: Feb 2007
Location: Luton, Bedfordshire, UK.
Posts: 456
|
Quad 303 problems
Hi All
I bought this amp as a complete wreck. It looked like it's been in a barn, top cover completely rusted, missing bottom cover, broken driver boards, hacked mains wiring. It was however cheap at £30 and I got a further fiver off when the seller realised the bottom cover was missing. Due to the broken driver boards I bought a set of 303 clone boards from eBay which I've populated. Powering it up on the lamp limiter for the first time I'm getting the lamp going out as usual but after a few seconds but then I'm getting medium bright pulsing every 2 seconds and driver board voltages between pins 1 & 9 are fluctuating between 30 odd and 70 odd volts dc. I've rebuilt the boards using new parts, replaced output and smoothing caps but haven't touched output transistors or the rectifier block. Clearly the amp is drawing excess current on a 2 second cycle. Does anyone have an idea what the best place to start de bugging is. Assume I half know what I'm doing! Thanks Mike Ps I will report back but there could be a delay I work long hours. |
28th Jun 2017, 7:42 pm | #2 |
Dekatron
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,991
|
Re: Quad 303 problems
There is a note in the 303 service manual that says "As this form of regulator is not self-starting R200, R201 and MR201 are included to ensure the terminal voltage never falls below half the input voltage"
It could be that there is a problem related to those components, or it is just freaking out as a result of the lamp in the mains supply. Craig |
28th Jun 2017, 8:05 pm | #3 |
Hexode
Join Date: Jun 2011
Location: Glossop, Derbyshire, UK.
Posts: 487
|
Re: Quad 303 problems
Hi here is the service and user manual for the 303 hope they help
Dave
__________________
Any tool can be used as a hammer but a screwdriver makes the best chisel |
28th Jun 2017, 9:29 pm | #4 |
Heptode
Join Date: Jul 2009
Location: Stoke-on-Trent, Staffordshire, UK.
Posts: 719
|
Re: Quad 303 problems
I'm presuming you have checked the output transistors and their wiring, and also checked that the mica insulators are still actually insulating them from the chassis (you mentioned corrosion, so it's a possibility the insulation is damaged), and that the lamp is something like a 100Watt one in your limiter.
Unfortunately, the "clone" pcb's might have been arranged for different transistor lead out to those you were expecting, or even have some other unpredictable idiosyncracies. You also mentioned using new parts... unless identical to Quad's originals these could also have different pin-outs (ie EBC versus ECB and so on). Double check for pin-out differences and even pnp and npn in the correct places... we've all done it at some time, so again worth checking if you haven't already. Then: First, make sure that RV101 is set for minimum quiescent current, which is at maximum resistance in both channels. Then retest and adjust as per Quad service info. [Thinking here is that if this is set very high the supply will be pulled down causing the lamp limiter to light up and the supply voltage to fall, then quiescent current will reduce and cycle will repeat.] If still no joy: Isolate both power amplifiers by lifting their feeds from pin 1. Switch on and check the regulated supply is adjustable on the PSU board, and set it to 67V between pins 12 and 13. Add a small load (bear in mind the input current is limited by the lamp, so say 2.2kohms 3W) and check it's still 67V. You can use a heavier load but if the lamp limiter starts to light up its own resistance will increase and you get back into chasing your tail. Switch off and discharge the reservoir capacitors or wait until the voltage decays. You should know now whether the PSU board is ok or not. There are many ways of skinning a proverbial cat, but if you have a variable DC supply with adjustable current limit (or one with say 50-100mA limit) I would now connect this external supply to each amplifier in turn. Slowly increase the applied voltage from zero. (If you don't have a variable power supply you can jump to the next section, but it might be harder on the output transistors if there's a fault.) The amplifiers should start to operate at a fairly low voltage and you can check that the midpoint voltage follows at half the applied supply and can be adjusted to half the applied voltage using RV100... this is a good indication that the circuit is functioning correctly. Check the quiescent current does not rise rapidly with supply voltage and doesn't keep on rising. Set this for 5-10mA at 67V, or for a bit less if you can only get up to say 48V.. but check it is adjustable and in the right ball park. If there is a sudden jump in current it might indicate oscillation. Do this for each channel to confirm basic functionality. Disconnect your external power supply and reconnect internal feed to one power amplifier. Check and adjust mid-point voltage at pin 5 to be at 33.5V. Increase quiescent current setting to 5 to 10mA as per Page3 of manual. If this causes the lamp limiter to light up, there's a fault somewhere causing excessive current, possibly even instability. Either way, then power up the second amplifer and isolate the first one... then repeat procedure. You might have one channel that works and one that doesn't or they might both be good or bad. If both are good, reconnect both supply feeds and you should be ready to go. Hope this helps, or at least helps to narrow down the fault/s. Pete |
28th Jun 2017, 9:48 pm | #5 |
Hexode
Join Date: Feb 2007
Location: Luton, Bedfordshire, UK.
Posts: 456
|
Re: Quad 303 problems
Thanks guys and especially Pete for the very thorough steer. I did carefully check the pinout of the replacement transistors using the data sheets and circuit diagram. I will lift the cover and check the insulators under the outputs then work my way through your suggested approach probably at the weekend!
Thanks Mike |
28th Jun 2017, 10:10 pm | #6 |
Heptode
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
|
Re: Quad 303 problems
A bit OT, but it blows me away how helpful you guys and gals are. I am not a tinkerer or a repair sort of guy (too busy listening and looking at vintage tech to go down that route), but I love the detailed responses as above, and I learn a good deal from these responses, so thanks.
All the best from Bill. |
29th Jun 2017, 9:10 am | #7 |
Hexode
Join Date: Oct 2010
Location: Box End, Beds. UK.
Posts: 271
|
Re: Quad 303 problems
Out of curiosity, does anyone know what the Quad transistor type number are in terms of device manufactures part numbers? I gather that 38494 = 2N3055, for example.
|
29th Jun 2017, 10:40 am | #8 |
Dekatron
Join Date: Jul 2008
Location: Kington, Herefordshire, UK.
Posts: 3,674
|
Re: Quad 303 problems
This has been gone into elsewhere, but the 38494 is an RCA device, as originally used on the 303. Although broadly similar to the contemporary 2N3055, it is of hometaxial construction and hence has a lower ft than more modern epitaxial equivalents. This can cause stability problems in the 303.
|
29th Jun 2017, 12:45 pm | #9 |
Dekatron
Join Date: Jun 2013
Location: East Sussex, UK.
Posts: 3,326
|
Re: Quad 303 problems
You can get a 2N3055H which I believe to be an 'old style' hometaxial type
|
30th Jun 2017, 12:58 pm | #10 |
Nonode
Join Date: Jun 2009
Location: Herefordshire, UK.
Posts: 2,495
|
Re: Quad 303 problems
By the way - examine carefully all the large electrolytic capacitors for electrolyte oozing or swelling of the tops (where the pin connectors are). These have a history of giving problems.
|
30th Jun 2017, 1:24 pm | #11 |
Dekatron
Join Date: Jul 2008
Location: Kington, Herefordshire, UK.
Posts: 3,674
|
Re: Quad 303 problems
The H is to do with the voltage rating, not the die. Nobody makes the 3055 of fifty years ago now, and old stock is getting harder to find. Getting rid of the oscillation hazard with faster devices is no picnic...
|
30th Jun 2017, 1:42 pm | #12 |
Dekatron
Join Date: Jun 2013
Location: East Sussex, UK.
Posts: 3,326
|
Re: Quad 303 problems
Thanks for the correction Ted. I was given the wrong info a few years ago.
|
30th Jun 2017, 2:11 pm | #13 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,894
|
Re: Quad 303 problems
As Ted says, the 'Quad Triplet' output transistor arrangements were pretty clever in their day but they relied on the TO3 power transistors being very much slower than the pair of transistors driving each one. They are little feedback loops and there are no deliberate time-constants regulating the loop gain/bandwidth/phase other than the silicon bits themselves.
If you fit faster transistors (and with 800kHz minimum Ft, anything newer is going to be faster) you can't bring the loops under control by adding capacitors because that just adds more phase shift and that of the new power transistor is still there. You haven't moved its pole, you've added an extra one. The usual symptom is the amp making bursts of oscillation at one particular level when you're driving an audio sinewave into a load. One cheat is to nick the real old 2N3055 often lurking in the power regulator position, but that only gets you one transistor. However, the regulator is more tolerant of modernity. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
30th Jun 2017, 6:42 pm | #14 |
Hexode
Join Date: Feb 2007
Location: Luton, Bedfordshire, UK.
Posts: 456
|
Re: Quad 303 problems
Hi Steve
I've replaced all the main electrolytics. Mike |
1st Jul 2017, 11:07 am | #15 |
Dekatron
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,991
|
Re: Quad 303 problems
The 40411 (used as the series pass regulator) has an fT of 800kHz. And the contemporary 1975 2N3055 for the 3003 also had an fT of 800kHz(min) with no max specified.
The weird thing is that there are currently two flavours of 2N3055 from ON-semi. The 2N3055 has fT of 2.5MHz(min) with no max specified and is £1.72 from Farnell. And the 2N3055AG, which has "a larger SOA than the 2N3055" has an fT of 800kHz(min) but 6MHz (max). This is £4.32 from Farnell. But their "Multicomp" brand do a 2N3055H with an fT of 800kHz (min) with no max specified. £2.84. |
1st Jul 2017, 12:36 pm | #16 |
Octode
Join Date: Jul 2011
Location: Staffordshire Moorlands, UK.
Posts: 1,479
|
Re: Quad 303 problems
DaDa Electronics, who specialise in old Quad stuff, recommend the MJ15003 for replacing output transistors in the 303. They report very few frequency issues. If you are replacing transistors make sure that you do them in pairs, i.e. use two of the same device in each channel, don't mix 'em up.
FWIW DaDa use the MJ802 in the power supply. Also seeing as we're at it, their little upgrade kits for the 33 preamp are great! Cheers, Steve. PS Also a very good idea to replace the preset resistors too as intermittent wiper contact on the aged skeleton types causes reliability problems.
__________________
Those who lack imagination cannot imagine what is lacking... Last edited by fetteler; 1st Jul 2017 at 12:39 pm. Reason: PS |
1st Jul 2017, 1:14 pm | #17 |
Hexode
Join Date: Jun 2011
Location: Glossop, Derbyshire, UK.
Posts: 487
|
Re: Quad 303 problems
I agree with Steve the DADA kit for the 33 is great . I did my own 33 with the kit but bought the parts for the other 5 from RS ( that is why I have 200 BC550 as min order was for 250 ) the upgrade is quite easy but the later Quad's I have found to have poor quality printed boards that tend to lift
Dave
__________________
Any tool can be used as a hammer but a screwdriver makes the best chisel |
18th Sep 2017, 9:58 am | #18 |
Moderator
Join Date: Jan 2003
Location: Ipswich, Suffolk, IP4, UK.
Posts: 21,289
|
Re: Quad 303 problems
__________________
Graham. Forum Moderator Reach for your meter before you reach for your soldering iron. |